Cantrip scaling

[Burley]

I'm sure this has been asked and answered a dozen times, but: Do cantrips scale on class or character level?

My wizard will be multiclassing to Warlock at level 5 (for story reasons). When I gain Eldritch Blast, will it be a "5th level cantrip" or a "1st level cantrip?"

[Aett_Thorn]

Based on Character level

[Zhorn]

To help with future such inquiries; level requirements will generally state if they are class specific or not, and generally that is only true of class features, not spells/cantrips.
Warlock's Eldritch Invocations feature for example specifies on the last line that the levels are Warlock levels.
Spells/Cantrips are not owned by any specific class, even if they might be unique to a classes spell list like with Eldritch Blast, so their level scaling is just off your total level (or caster level in the case of spellcasting monsters and NPCs)

[Chronos]

That's definitely the RAW. Personally, though, I prefer to houserule that cantrip improvement is a class feature of wizards, sorcerers, warlocks, bards, clerics, and druids (and I suppose artificers, if they're in use at your table), gained at 5th, 11th, and 17th levels. Which means that a wizard who multiclasses to warlock is in the same boat as a barbarian who multiclasses to fighter, or the like. And also means that Magic Initiate on a non-caster doesn't scale at all.

[Yuki Akuma]

That's definitely the RAW. Personally, though, I prefer to houserule that cantrip improvement is a class feature of wizards, sorcerers, warlocks, bards, clerics, and druids (and I suppose artificers, if they're in use at your table), gained at 5th, 11th, and 17th levels. Which means that a wizard who multiclasses to warlock is in the same boat as a barbarian who multiclasses to fighter, or the like. And also means that Magic Initiate on a non-caster doesn't scale at all.

So someone who gets a cantrip from their race or a feat just has it become totally useless over time?

[Chronos]

Yup, just like someone who picks up a weapon proficiency from a race or feat has it become less useful over time. Though not totally useless, in either case: For instance, for a paladin, even a non-scaling eldritch blast is probably still the best ranged option they have available, and anyone can make use of setting something on fire at a distance.

[NaughtyTiger]

That's definitely the RAW. Personally, though, I prefer to houserule that cantrip improvement is a class feature of wizards, sorcerers, warlocks, bards, clerics, and druids (and I suppose artificers, if they're in use at your table), gained at 5th, 11th, and 17th levels. Which means that a wizard who multiclasses to warlock is in the same boat as a barbarian who multiclasses to fighter, or the like. And also means that Magic Initiate on a non-caster doesn't scale at all.

I like this a lot.



So someone who gets a cantrip from their race or a feat just has it become totally useless over time?

only a if they chose a combat cantrip

[tiornys]

That's definitely the RAW. Personally, though, I prefer to houserule that cantrip improvement is a class feature of wizards, sorcerers, warlocks, bards, clerics, and druids (and I suppose artificers, if they're in use at your table), gained at 5th, 11th, and 17th levels. Which means that a wizard who multiclasses to warlock is in the same boat as a barbarian who multiclasses to fighter, or the like. And also means that Magic Initiate on a non-caster doesn't scale at all.
Paladins who take Blessed Warrior and Rangers with Druidic Warrior should get cantrip scaling as part of taking that fighting style.

[KorvinStarmast]

Paladins who take Blessed Warrior and Rangers with Druidic Warrior should get cantrip scaling as part of taking that fighting style. This is a reasonable position. +1

[Psyren]

Uh... what about racial cantrips? Do those just never improve with this houserule, even though the character's ability with that cantrip is due to racial magic and not due to class training?

[sithlordnergal]

They improve with character level, and I don't really see much of a need to houserule it to only work with class level. Cantrips aren't really that impressive, and even Eldritch Blast is only so-so if you don't have the Invocations to support it. Its better than a javelin for sure, but I chalk that up to javelins being bad, not cantrips being good.

[PhoenixPhyre]

They improve with character level, and I don't really see much of a need to houserule it to only work with class level. Cantrips aren't really that impressive, and even Eldritch Blast is only so-so if you don't have the Invocations to support it. Its better than a javelin for sure, but I chalk that up to javelins being bad, not cantrips being good.

Yeah. And even with Agonizing Blast, EB still isn't that great. Better than other cantrips, but you spent two levels to get it.

[sithlordnergal]

Yeah. And even with Agonizing Blast, EB still isn't that great. Better than other cantrips, but you spent two levels to get it.

That, or spent a level and a feat, which is even worse. XD Honestly, there are only two cantrips that I feel stand out above the rest. The first is Booming Blade due to it being a weapon attack, and therefore can work with things like Sneak Attack and Smite. The second is Chill Touch, not due to damage, but due to the fact that stopping healing is a critical niche that can be insanely important when it comes up.

[NaughtyTiger]

The second is Chill Touch, not due to damage, but due to the fact that stopping healing is a critical niche that can be insanely important when it comes up.

This is a great reason why cantrips shouldn't scale.
most damage cantrips have riders that offset the few points of damage from the ability modifier,

chill touch - healing/damage type
viscous mockery - disadvantage/damage type
thorn whip - auto-pull
acid splash/green flame blade - 2nd target
booming blade - mobility control
.... [saves instead of AC]

flexibility should reduce damage

[sithlordnergal]

This is a great reason why cantrips shouldn't scale.
most damage cantrips have riders that offset the few points of damage from the ability modifier,

chill touch - healing/damage type
viscous mockery - disadvantage/damage type
thorn whip - auto-pull
acid splash/green flame blade - 2nd target
booming blade - mobility control
.... [saves instead of AC]

flexibility should reduce damage

See, the thing is most cantrips aren't really flexible, and those that have an amazing rider do a lot less damage. The only exception to this is Chill Touch, which I consider to be the strongest cantrip in the game. Even more so than Booming Blade or Eldritch Blast. Looking at the list you provided:


Chill Touch - Fully agree its OP. It deals d8 damage, stops all healing, deals Necrotic damage, and if you hit an undead they have disadvantage on all attack rolls against you for the round. Though ironically its at its strongest and weakest when used against Undead. Still, only cantrip that's actually OP.
Vicious Mockery - Disadvantage/Damage Type yes, but only deals d4's, and the Disadvantage is only on a single attack, not a round. Could use a small buff actually
Thorn Whip - Auto-Pull and best damage type, but is only 30ft, deals d6's, can only pull Large or smaller. This one is well balanced
Acid Splash - Good damage type, nothing else. It can target a 2nd creature within 5ft of the target, but honestly needs a buff
Green Flame Blade - Bad damage type, makes up for it with the benefits of being a weapon attack. Pretty good spell to be honest, I think its balanced
Booming Blade - Great damage type, has the benefits of being a weapon attack, great damage, handy control rider. I could see this being nerfed to a d6 cantrip, but its not really needed.
Saves instead of AC - Keeps the casters useful in encounters against things with high AC, since they usually can't grapple/shove/ect. like Martials can. Plus they deal no damage on a save. That's perfectly fine


With the exception of Booming Blade and Chill Touch, the cantrips you listed have a good balance between damage and utility. And even then Booming Blade is only marginally better than the others, while Chill Touch is the only cantrip that is OP. The stronger ones, such as Vicious Mockery, deal almost no damage for their special effect. Meanwhile, cantrips like Firebolt deal a ton of damage, but have very limited utility. After all, when was the last time you used Firebolt's ability to set flammable objects that aren't being worn or carried on fire?

By having them scale with character level instead of class level, they're always going to be handy in niche situations, but the classes that worry about character vs class levels are going to have so many better things to do. Even a Soradin with Eldritch Blast+Agonizing Blast is going to prefer being in melee with an enemy, because their melee attacks are going to be so much better than EB at all levels.

The only times cantrips do outperform everything is when you compare cantrips to thrown weapons...but that's not an issue with cantrips being too strong, its an issue with thrown weapons being garbage. The answer to cantrips shouldn't be "make them as bad as thrown weapons", it should be "buff thrown weapons to be on par with literally everything else"

[Psyren]

flexibility should reduce damage

It does. The most damaging cantrips (biggest dice) have no riders.

[NaughtyTiger]

The stronger ones, such as Vicious Mockery, deal almost no damage for their special effect

3 points:

1 they ALL deal magical damage
b most deal non-BPS damage
iii a cantrips [zero-resource action] with riders shouldn't also do damage. Casters can already do control and plenty of damage: single target, AOE, against AC or many Save options. If they burn all of their BAMF options, then they already shined. Help the others shine by doing riders.

[tiornys]

a [zero-resource action] with riders shouldn't also do damage
A little out of context, but man do I disagree strongly with this. The solution is not to remove riders from cantrips, but to give martials more ways to add at-will riders to their attacks (see for example, Crusher).

[LibraryOgre]

A little out of context, but man do I disagree strongly with this. The solution is not to remove riders from cantrips, but to give martials more ways to add at-will riders to their attacks (see for example, Crusher).

That lack is part of what I think leads to martials being seen as boring. If I hit you, I should have a chance to knock you down. Or push you back. Or use the opening to take a step back myself.

One of the reasons 4e worked is because martial attacks were more than "1d8+Str damage". The Basic Attack? That was the boring thing that should never happen.

[sithlordnergal]

3 points:

1 they ALL deal magical damage
b most deal non-BPS damage
iii a cantrips [zero-resource action] with riders shouldn't also do damage. Casters can already do control and plenty of damage: single target, AOE, against AC or many Save options. If they burn all of their BAMF options, then they already shined. Help the others shine by doing riders.


1) Yes, and?

2) Your point here? Non-BPS damage doesn't tend to matter that much. It can let you bypass Resistance/Immunity to non-magical BPS, but depending on the damage type it could be resisted or ignored just as much

3) Sooo...do you expect casters to just sit back and do nothing but damage or nothing at all once their slots are gone? With the exception of Chill Touch, none of the riders are all that powerful.


There's a reason casters stop using cantrips the moment they have more than 3 or 4 spell slots per day/encounter. Cantrips are not a good use of one's action. They are the thing you fall back on when you have 0 resources left. Its something that lets them continue to effectively contribute when they're all out of resources. Martials don't actually have this issue. Even if they're all out of resources, a Martial can still effectively contribute via the attack action. Once people have Extra Attack, casters without it cannot. Cantrips are used to make up for the lack of Extra Attack, without being as good as Extra Attack.

Cantrips with riders already scale their damage down based on the rider, and with the exception of Chill Touch, which could have its damage lowered to a d6, and Acid Splash, which needs a damage buff to d8, they're all pretty well balanced and flavorful. They have the riders to help you still feel like you're playing a caster without being so strong that they end up better than 1st level spells. And removing the damage or riders entirely would be a terrible idea because:

A) You'll either split them into separate cantrips and now all cantrips deal d10 damage and feel exactly the same, while also straining an already very limited resource that's difficult to change

or

B) You just make it so casters can't do anything once their spell slots are gone by removing damage entirely, which has the add on effect of enforcing the 5-minute Adventuring Day cause casters will always want a long rest the moment they start to get low on slots even more than they do now.

[NaughtyTiger]

1) [Cantrips do magic damage] Yes, and?

2) Your point here? Non-BPS damage doesn't tend to matter that much. It can let you bypass Resistance/Immunity to non-magical BPS, but depending on the damage type it could be resisted or ignored just as much

3) Sooo...do you expect casters to just sit back and do nothing but damage or nothing at all once their slots are gone? With the exception of Chill Touch, none of the riders are all that powerful.


1) Swords don't do magic damage without DM fiat. So d8 magic damage is better than 0*4*(d12 + 4) non-magic damage when dealing with a critter immune to non-magical damage.

2a) " It can let you bypass Resistance/Immunity to non-magical BPS," yep. that is the point of magical damage. thanks for explaining it
2b) "depending on the damage type it could be resisted or ignored" yep. thank goodness, a typical caster starts with 3+ cantrips, so 3+ damage types if the enemy is immune to 2 of them...

3) sooo.. you expect martials to just sit back and do nothing but damage or nothing at all cuz they never had spell slots to start with...




There's a reason casters stop using cantrips the moment they have more than 3 or 4 spell slots per day/encounter. ... They are the thing you fall back on when you have 0 resources left. ...
Martials don't actually have this issue. Even if they're all out of resources, a Martial can still effectively contribute via the attack action.


I agree with this, and this is the problem.
While casters have slots, they can out perform a fighter. They can do more single target damage. They can do more AoE damage. They can knock out a 180HP baddie in 1 round.
And with scaling cantrips+riders they can still keep up with the fighter.... there is no penalty for going all out on a fight early.



.... They have the riders to help you still feel like you're playing a caster without being so strong that they end up better than 1st level spells....

you just told me that "casters stop using cantrips the moment they have more than 3 or 4 spell slots"




A) You'll either split them into separate cantrips and now all cantrips deal d10 damage and feel exactly the same, while also straining an already very limited resource that's difficult to change
B) You just make it so casters can't do anything once their spell slots are gone by removing damage entirely,

A) d10 fire damage is very different from d10 thunder damage which is very different that d10 radiant damage...
whereas d10+5 bludgeoning is the same as d10+5 slashing, same as d10+5 piercing

B) if "imposing disadvantage on an attack" or "separating a downed friendly from her enemy" is doing nothing then, we are playing different games



A little out of context, but man do I disagree strongly with this. The solution is not to remove riders from cantrips, but to give martials more ways to add at-will riders to their attacks (see for example, Crusher).
Crusher is a [half] Feat. A Feat is definitely not a zero-cost resource. If you need a Feat (or an Invocation or metamagic) to add a rider to a single spell, I am okay with that.

[Psyren]

Base Unarmed Strike now has Shove and Grapple built into it, so that gives us a bit of an indication of where they're headed. The martial classes themselves may have additional riders built in.

[NaughtyTiger]

Base Unarmed Strike now has Shove and Grapple built into it, so that gives us a bit of an indication of where they're headed. The martial classes themselves may have additional riders built in.

do you mean that an attack using unarmed strike can be replaced with a shove attack or a grapple attack instead of damage?

[Psyren]

do you mean that an attack using unarmed strike can be replaced with a shove attack or a grapple attack instead of damage?

Yes. Moreover, you can do both shove + damage if you take Tavern Brawler.

[sithlordnergal]

1) Swords don't do magic damage without DM fiat. So d8 magic damage is better than 0*4*(d12 + 4) non-magic damage when dealing with a critter immune to non-magical damage.

2a) " It can let you bypass Resistance/Immunity to non-magical BPS," yep. that is the point of magical damage. thanks for explaining it
2b) "depending on the damage type it could be resisted or ignored" yep. thank goodness, a typical caster starts with 3+ cantrips, so 3+ damage types if the enemy is immune to 2 of them...

3) sooo.. you expect martials to just sit back and do nothing but damage or nothing at all cuz they never had spell slots to start with...


1) It is true that swords don't do magic damage, but that's generally fixed by handing out a magic weapon. Not to mention there are now three classes that can net you a guaranteed magical weapon, Blade Warlock, Forge Cleric, and Artificer. Additionally, should we strip the Monk of their automatic magical damage at level 6 just because magic damage is better than a non-magical sword?

2) Typical caster starts with 3 cantrips or less actually, and while they can choose to only have damaging cantrips, this is an extremely limited resource. Heck, up until Tasha's or Xanathar's, they couldn't change their cantrips at all.

3) Martials are able to do so much more than just damage. If you think that martials can only deal damage, then you're not very good at playing a martial. Obviously you have the standard grapple and shove, which are great control options on their, they can also disarm, tumble through an enemy space or overrun them, tank. If you're a Rogue, your subclass might give you multiple options, Fighter subclasses tend to have additional things they can do, same with Rangers, Monks, and Paladins. Now, while casters can technically try those things as well, lets be honest. They do not have the tools to do so, not like a martial can.



I agree with this, and this is the problem.
While casters have slots, they can out perform a fighter. They can do more single target damage. They can do more AoE damage. They can knock out a 180HP baddie in 1 round.
And with scaling cantrips+riders they can still keep up with the fighter.... there is no penalty for going all out on a fight early.

you just told me that "casters stop using cantrips the moment they have more than 3 or 4 spell slots"


So, while casters are better than martials when they have slots, cantrips are a far cry from an Extra Attack. The scaling makes them better then throwing a Dagger or firing a single shot from a Light Crossbow, but they're not as effective as a Fighter with Extra Attack. Where is where the penalty of using all your resources early comes in. Its why casters tend to stop using cantrips once they have enough slots to go through an entire adventuring day, but can still rely on them.

Not only that, but the casters that tend to use cantrips the most are low level casters without the spell slots to last an entire adventuring day. Low level casters should be allowed to still feel like a caster via their at-will abilities, just like Rogues feel like Rogues and Fighters feel like Fighters.



A) d10 fire damage is very different from d10 thunder damage which is very different that d10 radiant damage...
whereas d10+5 bludgeoning is the same as d10+5 slashing, same as d10+5 piercing

B) if "imposing disadvantage on an attack" or "separating a downed friendly from her enemy" is doing nothing then, we are playing different games

A) No, there is fundamentally no difference. Though I do agree that there's also no difference between BPS. But that's an issue across all damage types, and 5e should have done a better job at making the different damage types matter outside of just elemental resistances.

B) Depending on the level, imposing disadvantage on a single attack roll is basically doing nothing. Its fine at low levels, but when creatures consistently have 2 to 3 attacks, disadvantage on one attack is nothing. Its why Vicious Mockery deals some d4's of damage.

[TaiLiu]

3 points:

1 they ALL deal magical damage
b most deal non-BPS damage
iii a cantrips [zero-resource action] with riders shouldn't also do damage. Casters can already do control and plenty of damage: single target, AOE, against AC or many Save options. If they burn all of their BAMF options, then they already shined. Help the others shine by doing riders.

I think non-BPS damage is sometimes overrated. Equal to or better than non-magical BPS? Sure. Better than magical BPS? No way. Magical BPS is the best. I feel like spells that do magical BPS damage ought to be changed so that magical BPS can be the niche of martial classes.


3) Martials are able to do so much more than just damage. If you think that martials can only deal damage, then you're not very good at playing a martial. Obviously you have the standard grapple and shove, which are great control options on their, they can also disarm, tumble through an enemy space or overrun them, tank. If you're a Rogue, your subclass might give you multiple options, Fighter subclasses tend to have additional things they can do, same with Rangers, Monks, and Paladins. Now, while casters can technically try those things as well, lets be honest. They do not have the tools to do so, not like a martial can.

I don't think most martial classes can do most of those things, actually. Not unless a DM elects to use optional rules.

[Chronos]

Quoth sithlordnergal:

Low level casters should be allowed to still feel like a caster via their at-will abilities, just like Rogues feel like Rogues and Fighters feel like Fighters.
Poor example, since rogues don't get to feel like rogues until 11th level.

And while martials can do a few things other than damage, those things are utterly insignificant compared to what casters can do. Outside of houserules, what they can do is shove and grapple, and that's it. You need houserules for them to disarm or tank, and moving through an enemy's space isn't really doing anything.

[Burley]

I reject the idea that cantrips should scale with class rather than character. One of the reasons I'm dipping Warlock is that my Conjuration Wizard (built for pirate ship repairs) runs out of combat spells quickly. Monster stats will scale up assuming I can deal 2d6-2d10 (if I hit) and the other 3-4 people in my group will be doing the same.

I also reject the idea that martial characters and casters need to be held to similar standards, i.e. a cantrip shouldn't deal damage and have an effect. The roles of spell casters are more nuanced that most martial classes and dealing damage isn't their only job; having a rider effect on a spell that imposes disadvantage, penalties or removes the monster's ability to use reactions are a part of the caster's role in the group. Weaken the enemy and make it safer or more advantageous for other characters to do their jobs.


Poor example, since rogues don't get to feel like rogues until 11th level.

So, Reliable Talent is what makes a rogue? Why not Cunning Action letting them feel like a rogue? Why not Sneak Attack and the many ways the subclasses allow SA to happen? That seems like an arbitrary reference point for your argument. Making skill checks is a part of playing a Rogue, but it's not their defining role in a party.

[NaughtyTiger]

sithlordnergal, you are making my points very well, thank you
i will go through each point that I see as relevant, but this is making for very long posts.


1) It is true that swords don't do magic damage, but that's generally fixed by handing out a magic weapon.
we agree that a fighter needs special permission from the DM to play as equals.


now three classes [caster archetypes] that can net you a guaranteed magical weapon, Blade Warlock, Forge Cleric, and Artificer.
we agree that would be casters using their resources to help martials shine, unless the casters used it to help themselves in combat.


Additionally, should we strip the Monk of their automatic magical damage at level 6 just because magic damage is better than a non-magical sword?
of course not, thank goodness no one suggested that. why would we take away from a marital's limited ability. besides, this supports my point, doing magical damage is a class ability based on class levels, not PC level.


2) Typical caster starts with 3 cantrips or less actually, and while they can choose to only have damaging cantrips, this is an extremely limited resource.
so cantrips don't need to do damage to be useful.



3) Martials are able to do so much more than just damage. If you think that martials can only deal damage, then you're not very good at playing a martial. Obviously you have the standard grapple and shove, which are great control options on their, they can also disarm, tumble through an enemy space or overrun them, tank.
quick! how much damage does a martial do during: grapple, shove, disarm, tumble, overrun...
almost like they are doing the rider without the damage


Now, while casters can technically try those things as well, lets be honest. They do not have the tools to do so, not like a martial can.
dude? they can... and more importantly, they shouldn't.


casters are better than martials when they have slots yes, casters are much better than martials when the have resources
so if you make them as good as martials when even when they don't have resources, then you are defining that casters are better than martials



the casters that tend to use cantrips the most are low level casters without the spell slots to last an entire adventuring day.
again, you are making my point very well.
low level casters don't have cantrip scaling yet.
additionally, low level martials only have 1 attack; so they get to do damage OR do a rider-effect (like shove).


No, there is fundamentally no difference [between fire and thunder damage]. Though I do agree that there's also no difference between BPS.
Abilities and resistances of trolls, devils, ... suggest otherwise


Depending on the level, imposing disadvantage on a single attack roll is basically doing nothing. Its fine at low levels, but when creatures consistently have 2 to 3 attacks, disadvantage on one attack is nothing.
Then imposing advantage on a single attack is basically doing nothing.


You keep saying that casters need to feel like casters even without resources.
I agree, but we disagree on what that means.

You appear to say they should be all powerful with spells during all 3 pillars of DnD. And still really strong during all 3 pillars after burning all of their spells.
You appear to suggest that a 2 level dip in Warlock should have the same damage output as a single class fighter

My view of casters is that they should be all powerful with spells in 1 of the 3 pillars of DnD (whichever they focus on), and once they have shined brightly, they should step back and support the others

[PhantomSoul]

In one group we used whichever is great of the number of levels in your class and the number of levels since you got the cantrip. Worked well, and definitely better than picking up a cantrip from Magic Initiate or multiclassing and suddenly being the bestest with it.

But I'll admit to wishing a lot more were based on class levels and dump so much being based on character levels.


I reject the idea that cantrips should scale with class rather than character. One of the reasons I'm dipping Warlock is that my Conjuration Wizard (built for pirate ship repairs) runs out of combat spells quickly.


That would seem more like an argument that it should scale with class rather than character level...

[NaughtyTiger]

The roles of spell casters are more nuanced than most martial classes, and dealing damage isn't their only job;

To summarize, casters are different from martials, because:


casters are supposed to do lots of damage AND apply riders
martials are supposed to do lots of damage


That is just bad game design

[PhoenixPhyre]

To summarize, casters are different from martials, because:


casters are supposed to do lots of damage AND apply riders
martials are supposed to do lots of damage


That is just bad game design

There are certainly bad things about caster design. Cantrip scaling is not one of them.

Consider an evocation wizard spamming firebolt (one of the highest-damaging cantrips). Relative to a shortbow-wielding rogue who never has advantage but has a dependable melee ally (or sometimes has advantage and sometimes doesn't get sneak attack), the cantrips are doing...about half of the rogue's damage. With no rider. A d8 cantrip is dealing about 40%. A d4 cantrip is doing basically nothing. EB + AB does more, but has a substantial investment to get there. The blade cantrips require investment into a non-casting stat AND kinda suck for damage unless you're regularly triggering the secondary damage. And even then they're just on par.

And those riders? Unless you're a full warlock with lots of invocations spent on EB, they're meaningless. The best of them (chill touch's healing reduction) only works against a tiny fraction of the MM (healing abilities on monsters are really really rare unless you're spamming priests and acolytes). Vicious mockery's disadvantage on the next attack falls off as soon as the target has more than one attack.

Cantrips are stronger than a single, off-main-stat, no boosts, weapon attack. To be sure. But they're a consolation prize. Focus on the real issues here, which is that casters have
a) too many slots
b) too many spells
c) each of which is too powerful
d) and concentration is trivial to become almost immune to having break on you.

[Captain Cap]

In 3.5 there aren't at-will scaling cantrips, and yet, the high level gap between casters and martial is far greater than in 5e. If there is a caster supremacy, it isn't the fault of cantrips, that's for sure.

[NaughtyTiger]

There are certainly bad things about caster design. Cantrip scaling is not one of them.

Consider an evocation wizard spamming firebolt (one of the highest-damaging cantrips). Relative to a shortbow-wielding rogue who never has advantage but has a dependable melee ally (or sometimes has advantage and sometimes doesn't get sneak attack), the cantrips are doing...about half of the rogue's damage. With no rider. A d8 cantrip is dealing about 40%. A d4 cantrip is doing basically nothing. EB + AB does more, but has a substantial investment to get there. The blade cantrips require investment into a non-casting stat AND kinda suck for damage unless you're regularly triggering the secondary damage. And even then they're just on par.

And those riders? Unless you're a full warlock with lots of invocations spent on EB, they're meaningless. The best of them (chill touch's healing reduction) only works against a tiny fraction of the MM (healing abilities on monsters are really really rare unless you're spamming priests and acolytes). Vicious mockery's disadvantage on the next attack falls off as soon as the target has more than one attack.

Cantrips are stronger than a single, off-main-stat, no boosts, weapon attack. To be sure. But they're a consolation prize. Focus on the real issues here, which is that casters have
a) too many slots
b) too many spells
c) each of which is too powerful
d) and concentration is trivial to become almost immune to having break on you.

1) why is the evocation wizard spamming firebolt and not using her spell slots?
2) how well is the rogue doing when the target is behind partial cover compared to the wizard popping toll for the dead? especially if the rogue's partner IS the partial cover.
3) magical damage IS a rider (level 6 martial feature), fire damage IS a rider
4) the rogue NEEDS a dependable ally, this enforces teamwork. Good game design feature.
5) a caster SHOULD do less damage than a martial if she is using at-will abilities, because they CAN do so much more with resource abilities.

multiple scaling cantrips with riders is a sign of "too many slots", "too many spells", "each of which is too powerful"

[Keltest]

1) why is the evocation wizard spamming firebolt and not using her spell slots?
2) how well is the rogue doing when the target is behind partial cover compared to the wizard popping toll for the dead? especially if the rogue's partner IS the partial cover.
3) magical damage IS a rider (level 6 martial feature), fire damage IS a rider
4) the rogue NEEDS a dependable ally, this enforces teamwork. Good game design feature.
5) a caster SHOULD do less damage than a martial if she is using at-will abilities, because they CAN do so much more with resource abilities.

multiple scaling cantrips with riders is a sign of "too many slots", "too many spells", "each of which is too powerful"

1: To conserve spell slots or because she's out of them I suppose. Why does anybody ever use cantrips?
2: Are we using feats? If so, then probably fine. Sharpshooter is wildly powerful. Even beyond that, attack roll cantrips suffer from cover too, whats your point?
3: Granted, but martials have plenty of ways to get that rider too. Its not unique to spellcasters.
4: Not really. They can just bonus action get advantage for sneak attack since Tasha's. Or attack from stealth.
5: In specific circumstances sure, but its actually pretty hard for a high level spellcaster to keep up with the per-round damage of a high level martial over several encounters. They only get a single digit number of combined spell slots above about level 5, and thats where the really big dumb single target damage spells that would push their burst above a martial come online. Theyre better at clearing fields of goblins, but worse at slaying dragons.

[Captain Cap]

multiple scaling cantrips with riders is a sign of "too many slots", "too many spells", "each of which is too powerful"
They are in no way "too powerful", with respect to the actual powerful stuff casters are capable of.

[PhantomSoul]

They are in no way "too powerful", with respect to the actual powerful stuff casters are capable of.

That doesn't really work as an argument; if already being too good is the justification for keeping extra things that make them more too good, it's just completely abandoning any idea that it shouldn't be completely broken.

[Captain Cap]

That doesn't really work as an argument; if already being too good is the justification for keeping extra things that make them more too good, it's just completely abandoning any idea that it shouldn't be completely broken.
The argument is that if you want to fix casters, you have to focus on other things than cantrips. By removing scaling from cantrips you just nerf one of the few ways non-pure casters can use magic without much investment, not the guys with dozens of spell slots.

[PhantomSoul]

The argument is that if you want to fix casters, you have to focus on other things than cantrips. By removing scaling from cantrips you just nerf one of the few ways non-pure casters can use magic without much investment, not the guys with dozens of spell slots.

Except that doesn't really make sense -- if cantrips are also a problem, then also fix cantrips.

[Captain Cap]

Except that doesn't really make sense -- if cantrips are also a problem, then also fix cantrips.
It does, because cantrips are not a problem, not for me and many others, at least.

[PhantomSoul]

It does, because cantrips are not a problem, not for me and many others, at least.

Ah, it seems I misread the implication of your comment though -- it seemed to suggest you did find them too powerful, just not the most egregious issue and therefore not the priority:


They are in no way "too powerful", with respect to the actual powerful stuff casters are capable of.

(Though I would admit to having issues with cantrips. Definitely character-level scaling should be dumped IMO, but I also wish they were more restricted both for balance and for flavour. I play spellcasters most and the cantrips are so unreasonably good that spending slots often feels pointless... and I get desirable riders on top of that. Dump them from most classes altogether, maybe except in a higher-magic variant [I do see the appeal of Easy D&D with spellcasters never actually caring about any stat but their spellcasting one and never actually seeing useful for their weapon proficiencies other than as worthless fluff]; sorcerers keep them. But maybe I'm just nostalgic for a version of the game I never actually played -- one in which spellcasters don't have a golden spoon right from level 1.)

[Captain Cap]

But maybe I'm just nostalgic for a version of the game I never actually played -- one in which spellcasters don't have a golden spoon right from level 1.)
I actually played such a caster, and honestly, yeah, he had his charm. You never knew when to burn that Magic Missile, so you mostly used your trusty crossbow and bad aim :smallbiggrin:

[animorte]

I actually played such a caster, and honestly, yeah, he had his charm. You never knew when to burn that Magic Missile, so you mostly used your trusty crossbow and bad aim :smallbiggrin:

Yup! I certainly remember playing a caster way back when the crossbow was my best friend. I thought, might as well be a Ranger. This shouldn’t be an issue, and the cantrips are fine. I believe that has in fact already been agreed upon though.

I think the most balanced we have spells are somewhere near the half-casters. Warlocks also have a solid balance in that aspect. I’ve said it many times by now; unfortunately we won’t see back tracking in power level because the majority has grown accustomed to the power creep that happens in almost every game everywhere at some point. They’ve been doing a better job at powering up the martial’s though, a bit.

[PhoenixPhyre]

1) why is the evocation wizard spamming firebolt and not using her spell slots?
2) how well is the rogue doing when the target is behind partial cover compared to the wizard popping toll for the dead? especially if the rogue's partner IS the partial cover.
3) magical damage IS a rider (level 6 martial feature), fire damage IS a rider
4) the rogue NEEDS a dependable ally, this enforces teamwork. Good game design feature.
5) a caster SHOULD do less damage than a martial if she is using at-will abilities, because they CAN do so much more with resource abilities.

multiple scaling cantrips with riders is a sign of "too many slots", "too many spells", "each of which is too powerful"

1. Because I'm comparing cantrips? Saying "well, but leveled spells..." is a red herring.
2. Still way better. WAY WAY WAY WAY better. Saves are generally less accurate than attacks, especially WIS saves.
3. Not really. Many, if not most, of the creatures resistant to non-magical BPS are also resistant or outright immune to fire.
4. Sure. But they also have tons of ways to get advantage, which makes the difference triple instead.
5. Great. So they are. No change necessary.

Scaling cantrips have nothing to do with too many slots or too many spells, each of which is too powerful. Because
a) cantrips don't cost slots
b) cantrips are chosen separately from leveled spells
c) cantrips aren't too powerful.

Edit: if you make cantrips not scale at all, no one will use the combat ones at all. They're utterly wasted space, even with the riders. Because damage is the best CC and all the riders really suck. You'd have to substantially beef up the riders (such as vicious mockery imposing disadvantage on everything for a full turn), which is exactly the wrong direction to go IMO.

[tiornys]

Cantrips are fine. The underwhelming level of control that a greatsword wielding martial exerts on their immediate surroundings is the #1 problem when comparing at-will capabilities of martials to at-will capabilities of casters.

[JNAProductions]

It is true, NaughtyTiger, that a DM can not give the martial players any magic weapons, but still throw enemies immune to non-magical weapons at them.
It is also true that a DM can throw nothing but fire-immune enemies at a 5th level Wizard who's only 3rd level offensive spell is Fireball and who's best cantrip for damage is Firebolt.

There's a potential issue with non-magic weapon resistant/immune foes and weapon-using classes lacking magical weapons... But have you ever seen it come up? Because this isn't something that's complicated to fix. It's as simple as giving a magic weapon.

Plus, not many monsters are outright immune to non-magical weapons. Resistance pops up all the time, but not immunity.

[PhoenixPhyre]

It is true, NaughtyTiger, that a DM can not give the martial players any magic weapons, but still throw enemies immune to non-magical weapons at them.
It is also true that a DM can throw nothing but fire-immune enemies at a 5th level Wizard who's only 3rd level offensive spell is Fireball and who's best cantrip for damage is Firebolt.

There's a potential issue with non-magic weapon resistant/immune foes and weapon-using classes lacking magical weapons... But have you ever seen it come up? Because this isn't something that's complicated to fix. It's as simple as giving a magic weapon.

Plus, not many monsters are outright immune to non-magical weapons. Resistance pops up all the time, but not immunity.

Even resistance is tied very strongly to a couple creature types. Fiends and celestials (who also are uniformly immune and/or resistant to several elemental damage types if they resist non-magical BPS, and devils often can by bypassed with silver), constructs (not immune to adamantine and have magic resistance if not immunity to a bunch of damage and condition types as well), and elementals. Oh, and incorporeal or boss-level undead (who often also resist necrotic, so there goes toll the dead). Other than that, the incidence drops way off.

Edit: the full breakdown by type, averaged over all CRs, taken from the "old" monster books--BPS I means "immune to non-magical BPS"; BPS R is "resistant to non-magical BPS" without regard to other bypass methods. The other columns don't break out resistance and immunity separately.


https://i.ibb.co/VmT9Twx/defenses.png (https://ibb.co/k1XBXmJ)

[NaughtyTiger]

Okay,
I get that I am alone in the at-will abilities shouldn't also cause damage.
I am surprised that I am alone in the at-will abilities shouldn't do more than 1 die of damage.

What should Viscous Mockery be to make it a cantrip worth taking?
Since Disadvantage on the next attack is worthless, and Wis saves are unreliable, Xd8 or Xd10?


Also, PheonixPhyre, where did you get that chart!? That is nice.



It is true, NaughtyTiger, that a DM can not give the martial players any magic weapons, but still throw enemies immune to non-magical weapons at them.
It is also true that a DM can throw nothing but fire-immune enemies at a 5th level Wizard who's only 3rd level offensive spell is Fireball and who's best cantrip for damage is Firebolt.

Player agency is a factor.
The martial doesn't CHOOSE to be given a magical weapon
The wizard CHOOSES 1 fire cantrip and 1 fire spell. Moreover, that means that wizard CHOSE 3 non-combat cantrips and 13 non-combat spells. I bet that would definitely help the party explore and social...

[JNAProductions]

Eldritch Knights can make their own magic weapons.
Players can choose to commission magic items, or quest for them, or trade, or many other things.

It’s not as easy as learning spells on level up, but it’s not impossible unless the DM makes it impossible. And if the DM both makes it impossible to get a magic weapon, and uses enemies that are immune to non-magical weapons… I’d say that’s a DM issue.

[NaughtyTiger]

1. Because I'm comparing cantrips? Saying "well, but leveled spells..." is a red herring.
2. Still way better. WAY WAY WAY WAY better. Saves are generally less accurate than attacks, especially WIS saves.
3. Not really. Many, if not most, of the creatures resistant to non-magical BPS are also resistant or outright immune to fire.
4. Sure. But they also have tons of ways to get advantage, which makes the difference triple instead.
5. Great. So they are. No change necessary.

Scaling cantrips have nothing to do with too many slots or too many spells, each of which is too powerful. Because
a) cantrips don't cost slots
b) cantrips are chosen separately from leveled spells
c) cantrips aren't too powerful.

Edit: if you make cantrips not scale at all, no one will use the combat ones at all. They're utterly wasted space, even with the riders. Because damage is the best CC and all the riders really suck. You'd have to substantially beef up the riders (such as vicious mockery imposing disadvantage on everything for a full turn), which is exactly the wrong direction to go IMO.

I have been fairly consistent that the reason I "feel" cantrips are too powerful is because they are packaged with spells and slots.
If you ignore 2/3 of a full caster's spellcasting, because it is a red herring, then ought to cut out 2/3 of the rogues sneak attack...


if you make cantrips not scale at all, no one will use the combat ones at all. They're utterly wasted space, even with the riders. Because damage is the best CC and all the riders really suck. You'd have to substantially beef up the riders
your argument is that if the combat cantrips were nerfed, then after casters burned their spell slots, they would just sit out combat rather than help their team.

[NaughtyTiger]

Eldritch Knights can make their own magic weapons.
Players can choose to commission magic items, or quest for them, or trade, or many other things.

It’s not as easy as learning spells on level up, but it’s not impossible unless the DM makes it impossible. And if the DM both makes it impossible to get a magic weapon, and uses enemies that are immune to non-magical weapons… I’d say that’s a DM issue.

eldritch knights are casters, cuz they have cantrips and spells
players can ask the DM to commission magic items, or quest for them, or trade, or many other things.

there are only 4 ways to deal magical damage:

DM grants it
player caster grants it
be a caster
select a class/archetype/feat that will grant it at a specific level

[PhoenixPhyre]

Also, PheonixPhyre, where did you get that chart!? That is nice.


The chart came from the Monster Math link in my signature. It's something I put together based on all the data from MM, Volo's, and MToF.


I have been fairly consistent that the reason I "feel" cantrips are too powerful is because they are packaged with spells and slots.
If you ignore 2/3 of a full caster's spellcasting, because it is a red herring, then ought to cut out 2/3 of the rogues sneak attack...


You: Cantrips with riders are too strong
Me: Cantrips suck
You: But what about leveled spells!
Me: that's a separate issue. Changing cantrips would break things and not fix the other issue. No more than changing your curtains fixes the fact that the roof has a hole in it. Cantrips aren't a balance factor one way or another. Fix the real problem.

And cantrips aren't always packaged with spell slots. Making them non-scaling means that anyone who gets, say, a racial damage cantrip now has a wasted feature. Because damage cantrips aren't worth it in T1 as they stand (the rider is too weak and a light crossbow is more damage and resistance is neglible), and without scaling they wouldn't be worth it in T2 or later (when they are marginally so now under some circumstances). You're literally always better off shooting a shortbow twice as someone with 0 DEX than casting a T1 cantrip once with your main stat. Including the riders.



your argument is that if the combat cantrips were nerfed, then after casters burned their spell slots, they would just sit out combat rather than help their team.

No, they'd (marginally-effectively) use weapon attacks instead. Just like they did in all the editions before. Because a wizard is always better off with a light crossbow than with a non-scaling firebolt. Yes, even against resistant creatures. That's how crappy cantrips are. They're better than a single weapon attack. Barely. But not much. Enough that they keep the fiction of "I use magic" alive without having any balance-related effect.

[sithlordnergal]

I don't think most martial classes can do most of those things, actually. Not unless a DM elects to use optional rules.

Fun fact, options like Tumble, Overrun, ect., aren't under optional rules. They're simply hidden in the DMG for some reason. The DMG is pretty clear about Optional Rules and Rule Variants, those options don't actually fall under either of those tags from what I could find.

[Psyren]

Fun fact, options like Tumble, Overrun, ect., aren't under optional rules. They're simply hidden in the DMG for some reason. The DMG is pretty clear about Optional Rules and Rule Variants, those options don't actually fall under either of those tags from what I could find.

I personally like these options but they are indeed optional.

"Dungeon Master's Workshop: This chapter contains optional rules that you can use to customize your campaign, as well as guidelines on creating your own material."
...
"Action Options: This section provides new action options for combat. They can be added as a group or individually to your game."

[sithlordnergal]

your argument is that if the combat cantrips were nerfed, then after casters burned their spell slots, they would just sit out combat rather than help their team.

For the amount they could contribute if the combat cantrips were nerfed? Yeah, it would be as if they were not there. It would be annoying to players, both team mates and the casters alike.

[NaughtyTiger]

You: Cantrips with riders are too strong

Um, no. "cantrips that deal damage are too strong" would be closer, but actually I would prefer
"Given spellcasting and multiclassing rules, cantrips with riders that also deal damage are too strong."



You: But what about leveled spells!
Me: that's a separate issue. Changing cantrips would break things and not fix the other issue. No more than changing your curtains fixes the fact that the roof has a hole in it. Cantrips aren't a balance factor one way or another. Fix the real problem.


If there are too many spell slots, and too many spells, and they are way toooooo powerful, then why would they use a cantrip?
The way I see it:

they burned all of their slots to shine already. martials don't need to play until the slots are used up.
cantrips are strong enough to compete that they don't need to use spell slots. that contributes to too many slots.

That is why I cannot separate cantrips from leveled spells.
So, I ask you to clarify, why a caster is using a cantrip in combat?


And I disagree that crossbow is better than rider. Disadvantage on a fire giant attack may prevent 29pts. Thorn whip forced move can prevent an opportunity attack.

[PhoenixPhyre]

Um, no. "cantrips that deal damage are too strong" would be closer, but actually I would prefer
"Given spellcasting and multiclassing rules, cantrips with riders that also deal damage are too strong."


The riders are meaningless. That's the whole point. They're either so situational as to be useless (no, most creatures don't have a way of healing) or don't have a big effect (imposing disadvantage on one attack is a really small thing past T1). And the bigger riders already have pitiful damage -- 4d4 flat is crap. That's 10 damage on average. That's less than a level 1 sneak attack; a light crossbow with a +4 DEX does almost the same amount of damage (9.5), has a way better hit chance, and has since level 1. Even 4d8 flat on an attack roll is only 18 damage. That's...pitiful. Especially since most of the features that buff cantrip damage don't apply to the ones with riders very well.

The reduced damage already more than makes up for the riders.



If there are too many spell slots, and too many spells, and they are way toooooo powerful, then why would they use a cantrip?
The way I see it:

they burned all of their slots to shine already. martials don't need to play until the slots are used up.
cantrips are strong enough to compete that they don't need to use spell slots. that contributes to too many slots.

That is why I cannot separate cantrips from leveled spells.
So, I ask you to clarify, why a caster is using a cantrip in combat?


False dichotomy. Generally, people use cantrips in combat to contribute some damage when any of
a) they're already low on resources relative to what is coming up (or what they think is coming up)
b) they don't have any spells that would contribute better (as a bard, this happens quite a lot due to concentration)
c) combat is almost over anyway--no point in spending a leveled spell on cleaning up the 3 injured goblins left (in which case only having riders doesn't actually do you any good).

None of those say that martials don't need to play as long as casters have slots or that cantrips compete. Cantrips are competing against a light crossbow. Against which they still aren't great.

And still doesn't address the issue that even removing cantrips entirely would not measurably change the overall power of casters vis a vis martials. Cantrips are a rounding error all the time. But they help people feel good about themselves without appreciably changing anything. Cantrip damage is, mostly, just a psychological trick.



And I disagree that crossbow is better than rider. Disadvantage on a fire giant attack may prevent 29pts. Thorn whip forced move can prevent an opportunity attack.

Death is the best CC. Taking out the fire giant a round earlier has an even better chance of preventing damage.

[TaiLiu]

Fun fact, options like Tumble, Overrun, ect., aren't under optional rules. They're simply hidden in the DMG for some reason. The DMG is pretty clear about Optional Rules and Rule Variants, those options don't actually fall under either of those tags from what I could find.

I personally like these options but they are indeed optional.

"Dungeon Master's Workshop: This chapter contains optional rules that you can use to customize your campaign, as well as guidelines on creating your own material."
...
"Action Options: This section provides new action options for combat. They can be added as a group or individually to your game."
Yeah, what Psyren said. I would absolutely love it if they were default abilities. But the only defaults martials have are attack, grapple, and shove.

[JNAProductions]

An Eldritch Knight is a caster secondary, martial primary.

And again-if a DM uses tons of non-magic resistant or immune monsters and does NOT give the weapon users a chance to get magic gear… is that a system issue? Or a DM issue?

[TaiLiu]

And again-if a DM uses tons of non-magic resistant or immune monsters and does NOT give the weapon users a chance to get magic gear… is that a system issue? Or a DM issue?
Arguably it's both. Magic items are optional. Monsters, including classics like lycanthropes, are found in the core MM. I can see an inexperienced DM, or a DM unfamiliar with the inner workings of 5e, deciding not to give magic items while having a party fight, say, werewolves.

I'm more inclined to blame the mechanics than a newbie DM. They shouldn't have at the same time optional magic items and a MM with BPS-immune critters.

[Keltest]

Arguably it's both. Magic items are optional. Monsters, including classics like lycanthropes, are found in the core MM. I can see an inexperienced DM, or a DM unfamiliar with the inner workings of 5e, deciding not to give magic items while having a party fight, say, werewolves.

I'm more inclined to blame the mechanics than a newbie DM. They shouldn't have at the same time optional magic items and a MM with BPS-immune critters.

It very much depends on the kind of game you want. There are rules for having your weapon silvered, which bypasses werewolf damage immunity as well. If you want a werewolf to be a horrible dangerous monster that takes a coordinated group effort to defeat, then having no magic weapons is great for that because primary spellcasters dont have a lot of great "kill it to death" single target damage options at that level and it really does become a game of tactics where the martials have to protect the wizards from a monster that they cant directly hurt.

[TaiLiu]

It very much depends on the kind of game you want. There are rules for having your weapon silvered, which bypasses werewolf damage immunity as well. If you want a werewolf to be a horrible dangerous monster that takes a coordinated group effort to defeat, then having no magic weapons is great for that because primary spellcasters dont have a lot of great "kill it to death" single target damage options at that level and it really does become a game of tactics where the martials have to protect the wizards from a monster that they cant directly hurt.
Yeah, absolutely. But that's why I feel that "no magic items" should be in the optional ruleset. Most games run with magic items, and if you want a no-magic-items game, you should plan for it explicitly and get everyone's buy-in.

[Chronos]

When the designers explicitly say that the game can be played just fine with no magic items at all, and that's not actually true, that's 100% the fault of the designers.

[Asmotherion]

Cantrips scaling with Character Level is one of the good things 5e introduced. Let's not go steps backwards, and instead focus on making things that are actually good.

[LibraryOgre]

Cantrips scaling with Character Level is one of the good things 5e introduced. Let's not go steps backwards, and instead focus on making things that are actually good.

I disagree. As others have said, it starts bringing them into combat parity with fighters, and starts equalling other warriors. While it puts a single point of failure (one attack roll) v. the several chances to succeed of a fighter making multiple attacks, some subclasses are still on par with cantrips that other warrior types are with weapons; for example, the Knowledge domain, which starts adding their Wisdom modifier to damage with cantrips; a 17th level fighter is making 3 d8+Str attacks with a long bow, but the cleric is making 4d8+Wisdom attacks with their Sacred Flame... when most fighty types are getting two attacks, the cleric is doubling their damage. And, then, the cleric will get another die (tripling their damage from 1st level) at 11th level, when martials (except fighters) aren't getting more attacks.

On top of this, the cleric has spells... they CAN do 4d8+Wisdom with their cantrip, or they can drop a pillar of fire on you and your friends, and you'll all take 4d6 damage, at least (if you save).

It contributes to the relative depowering of martials. While I agree that we should focus on making things that are actually good, that also means the core classes and mechanics.

[Keltest]

I disagree. As others have said, it starts bringing them into combat parity with fighters, and starts equalling other warriors. While it puts a single point of failure (one attack roll) v. the several chances to succeed of a fighter making multiple attacks, some subclasses are still on par with cantrips that other warrior types are with weapons; for example, the Knowledge domain, which starts adding their Wisdom modifier to damage with cantrips; a 17th level fighter is making 3 d8+Str attacks with a long bow, but the cleric is making 4d8+Wisdom attacks with their Sacred Flame... when most fighty types are getting two attacks, the cleric is doubling their damage. And, then, the cleric will get another die (tripling their damage from 1st level) at 11th level, when martials (except fighters) aren't getting more attacks.

On top of this, the cleric has spells... they CAN do 4d8+Wisdom with their cantrip, or they can drop a pillar of fire on you and your friends, and you'll all take 4d6 damage, at least (if you save).

It contributes to the relative depowering of martials. While I agree that we should focus on making things that are actually good, that also means the core classes and mechanics.

I disagree with your disagreement. If a fighter wants to do damage, he's going to be dealing potentially 8d6 +4xSTR +40 a round with just great weapon master and a regular greatsword, and it only gets worse from there. A rogue is going to be dealing their weapon of choice plus 10d6.

Cantrips are not threatening the fighter's damage niche.

[PhantomSoul]

I disagree with your disagreement. If a fighter wants to do damage, he's going to be dealing potentially 8d6 +4xSTR +40 a round with just great weapon master and a regular greatsword, and it only gets worse from there. A rogue is going to be dealing their weapon of choice plus 10d6.

Cantrips are not threatening the fighter's damage niche.

I love how the comparison fighter is getting a feat to compare to something that's a basic level 1 feature, and that fighter is a relevant pick because they're the only ones with extra attack scaling even vaguely like cantrips... but if the fighter wants to do something other than damage with that hit they need to sacrifice damage (e.g. to shove, very limited options) or have a generous DM (to get magic items -- again, to compare to a freaking cantrip cast by someone who also, you know, has levelled spells... and can get this level of power from one level -- not the fighter's 20 -- or a feat).

[Keltest]

I love how the comparison fighter is getting a feat to compare to something that's a basic level 1 feature, and that fighter is a relevant pick because they're the only ones with extra attack scaling even vaguely like cantrips... but if the fighter wants to do something other than damage with that hit they need to sacrifice damage (e.g. to shove, very limited options) or have a generous DM (to get magic items -- again, to compare to a freaking cantrip cast by someone who also, you know, has levelled spells... and can get this level of power from one level -- not the fighter's 20 -- or a feat).
ASI distribution is as much a class feature as anything else. But even without that, 8d6 +4xSTR is still bigger than 4d8+Casting stat, and thats the strongest that most damage cantrips scale.

And Rogues got a name drop too with their 10d6 sneak attack damage.

[animorte]

I agree that the cantrips are fine. They should absolutely level with character level. This should be a reliable option for martials to also have some form of access to, providing various forms of damage. Casters should also be allowed to have a limitless source of damage as well. The problem shows up when the casters have copious amounts of higher level stuff. Casters can also AoE damage, AoE crowd control, single target control, AND have the ability to make some skills obsolete with their spells.

I remember in 3.5e getting absolutely sick of using a crossbow all of the time as a caster just because I didn't have any other way to deal consistent damage. We house-ruled the cantrip scaling that is seen today. Might as well just play some form of archer.

I also think we should look to half-casters and pact-casting to solve this problem. Having fewer access to being able to do anything and everything you want with a spell list is a greater issue.

[LibraryOgre]

I disagree with your disagreement. If a fighter wants to do damage, he's going to be dealing potentially 8d6 +4xSTR +40 a round with just great weapon master and a regular greatsword, and it only gets worse from there. A rogue is going to be dealing their weapon of choice plus 10d6.

Cantrips are not threatening the fighter's damage niche.

If they have a feat, are three levels higher than the cleric, and are willing to take a -5 to all of their attacks, meaning they are hitting about 25% less than the cleric (more, if the cleric is targeting an unproficient save).

So, let's go with numbers.

The fighter, at 17th level, can manage 6d6 +15 (assuming a 20 Strength) + 30 (great weapon master) on every turn. Twice in a day, they can manage 12d6 + 30 + 60 (using action surge). However, if they want that extra 30/60 points, we have to multiply everything by .75, because they're taking a -5 to hit, and so are therefore 25% less likely to hit any given DC/AC. Their average is going to be about (21 + 20 + 30)*.75, or 53.25, or 106.5 on the two turns they can action surge. If they don't have the feat, it's 6d6+15, or 12d6+30; 36 and 72.

The cleric, with a cantrip, will be doing 4d8+5 at that level, which is about 23 damage if it's Sacred Flame. On a miss with Sacred Flame, they still do about 11 damage, rounded down. If they use Word of Radiance, they will do about 19 damage per target (3.5*4 + 5), with up to 8 targets, so 152 if they're surrounded. Word of Radiance doesn't have fail damage, though. If they're targeting a non-proficient save, they are 30% more likely to succeed (because they get to add their proficiency bonus, while the target does not, making it about 30% more likely the save fails IF the target also has +5 on that attribute; it might be as high as 60%, if its a non-proficient save against a -1 attribute modifier). Where the fighter is looking at multiple points where their attacks will fail, the cleric is looking at a single point, with about a 66% chance (sic attributes, most PCs have two strong saves) they'll have a +30-60% chance to succeed on that one roll.

The cleric can give a fighter a run for their money in cantrip v. weapon damage, under optimal circumstances, with the right spell... 23 v. 36 average is a fair gap, but even the fighter's 72 for an action surge round pales compared to 152 damage from a well-situated cleric. But, hey, we mentioned the fighter's action surge, right? What about the cleric's spells? Of which a cleric has 19, spread across 9 levels. Blade barrier? 6d10 slashing damage, to about 20 people. 33 damage * 20 (people in 5' spaces along the 100 length). 660 damage, 330 if everyone saves. They can do that 4 times a day, for up to 40 minutes... so, twice as often as the fighter can Action Surge, for a fair amount of time, not just a round. And they still have all their 1st-5th level spells.

And this is just the fighter! What about a ranger? If they're just using the attack action at 17th level, it's 2d8+2*Dex. If they're a Hunter, then it might be 16d8+16*Dex (using Volley fire), for an impressive 152 damage... so long as they have enough arrows. They can do just as much damage as a cleric can with Word of Radiance (and at a better, far safer, range, spread across more targets). Rangers have spells, which ups utility, but nowhere near the number of a cleric.

Paladin? Ideal circumstance, they can manage 2d6+6d8+5 on one hit (5th level smite, against an undead or fiend), and 5d8 on three more, then decreasing from there. Their ideal hit is 39; their ideal round is 73.5 (plus a bit and a rider if they use their bonus action and some lower level spell slots for Smite spells). Oath of Devotion might bring this up to 44 and 88, if they Channel Divinity. Lots of hits, though, they're going to be doing about 12 damage, or 24 for the round (Oath of Devotion can bring this up to 17 or 34 for a minute a day).

Clerics with cantrips aren't slouches in damage, and if we start including their spell damage, they become powerhouses of combat that a fighter can't hope to match.

[Keltest]

Sacred Flame doesnt have fail damage either. I dont think theres a single damaging cantrip that does, in fact. The way you phrased talking about proficient saves has also completely lost me. I have no idea what youre trying to communicate there.

I also cant help but notice that you accounted for potential misses with the fighter using GWM, but not potential saves made for Word of Radiance. You also seem confused about Action Surge, which is a short rest ability, not a long rest ability. And that's not even accounting for the fact that the cleric has to be in a white room scenario where being surrounded on all sides is not imminent death for that to be considered "optimal."

All of which is to say, I find these comparisons rather questionable.

ETA: On an additional reading, I figured out what you meant with the proficient saves.

[Mastikator]

Cantrips scaling with Character Level is one of the good things 5e introduced. Let's not go steps backwards, and instead focus on making things that are actually good.

Not only do I mega-agree, but I started bringing in cantrip wands into my game. It's just not a problem if players have access to fire bolt. I mean some cantrips are sacred cows (like eldritch blast) and won't come up, but ray of frost or thornwhip? Why not, it's fun.

[Psyren]

Cantrips scaling with Character Level is one of the good things 5e introduced. Let's not go steps backwards, and instead focus on making things that are actually good.

+1000


When the designers explicitly say that the game can be played just fine with no magic items at all, and that's not actually true, that's 100% the fault of the designers.

It can be, but they also set expectations for magic items by tier in the DMG. If you choose to not follow those guidelines, you need to make encounters easier to compensate.

[PhoenixPhyre]

+1000



It can be, but they also set expectations for magic items by tier in the DMG. If you choose to not follow those guidelines, you need to make encounters easier to compensate.

And even tells you as much in a sidebar. The system, in particular, does not assume you have weapons or armor that provides a +x to hit or ac. A moon touched blade (common) works just fine for resistance and immunity. If you go completely without magic items, then it suggests being careful about enemies resistant or immune to non magical attacks

[Chronos]

OK, so what they meant was that the game plays just fine without certain kinds of magic items. Then that's what they should have said.

[Captain Cap]

OK, so what they meant was that the game plays just fine without certain kinds of magic items. Then that's what they should have said.
The game plays fine without magic items in general. The DM is under no obligation to include enemies immune to mundane weapons in their game, they are DM Fiat as much as the availability of magic items. The only case you should worry about is if you have some sort of adversarial relationship with your DM, but that would be a problem on its own.

[Chronos]

Or if you have a DM sufficiently inexperienced that they think they can rely on the advice in the books and end up with a reasonable game.

[PhoenixPhyre]

Or if you have a DM sufficiently inexperienced that they think they can rely on the advice in the books and end up with a reasonable game.

You mean the advice to not use immune/resistant monsters if you're not giving magic weapons?

There is no advice in the books that says "don't give magic items" or even "you really don't need to give magic items." That's entirely fan-side. There is one sidebar in Xanathar's that says "you can get away without doing so, but be careful of..."

A few quotes:



The Dungeon Master’s Guide assumes a certain amount of treasure will be found over the course of a campaign. Over twenty levels of typical play, the game expects forty-five rolls on the Treasure Hoard tables, distributed as follows:

Seven rolls on the Challenge 0–4 table
Eighteen rolls on the Challenge 5–10 table
Twelve rolls on the Challenge 11–16 table
Eight rolls on the Challenge 17+ table
Because many of the table results call for more than one magic item, those forty-five rolls will result in the characters obtaining roughly one hundred items. The optional system described here yields the same number of items, distributed properly throughout the spectrum of rarity, while enabling you to control exactly which items the characters have a chance of acquiring.


And weapons are among the most common permanent magic items. Appearing way more frequently than armor, in fact. So yes, they're expected to have magic items.



ARE MAGIC ITEMS NECESSARY IN A CAMPAIGN?

The D&D game is built on the assumption that magic items appear sporadically and that they are always a boon, unless an item bears a curse. Characters and monsters are built to face each other without the help of magic items, which means that having a magic item always makes a character more powerful or versatile than a generic character of the same level. As DM, you never have to worry about awarding magic items just so the characters can keep up with the campaign’s threats. Magic items are truly prizes. Are they useful? Absolutely. Are they necessary? No.

Magic items can go from nice to necessary in the rare group that has no spellcasters, no monk, and no NPCs capable of casting magic weapon. Having no magic makes it extremely difficult for a party to overcome monsters that have resistances or immunity to nonmagical damage. In such a game, you’ll want to be generous with magic weapons or else avoid using such monsters.


So the rules assume someone has access to damage that bypasses resistance/immunity to non-magic weapons. Whether that's via spells, monks (the only non-casting class at the time this was written with innate access to magic weapons), or even NPCs capable of casting magic weapon. That's the bar.

Remember, this is a team game and no one should be capable of soloing relevant challenges. Which means casters need to be tuned down (because sometimes they can do so). Nerfing cantrip scaling is twiddling the radio knobs after the Titanic hit the iceberg. Fix the real issues, the low hanging fruit of
* polymorph, summoning spells generally, simulacrum, etc (aka the broken spells)
* generally too many spell slots (such that conservation over an average day isn't a big concern)
* spells that do too many things (so you can slot a single "do (almost) everything" swiss army knife, which makes prep limits not as binding)
* features (including multiclassing) that let casters play as martials almost just as well without expending significant daily resources doing so. Yes, that means hex warrior and bladesinger, as well as 1-level cleric (etc) dips need to go away. You should be paying roughly 2/3 of your casting to play at being a martial. Because full martials playing caster can only get to 1/3 caster status, and turnabout's fair play.
* etc.

Cantrips (other than maybe the weapon cantrips) and their scaling aren't even on this list. They're so far down it's not even funny.

[Eldariel]

Cantrips scaling with Character Level is one of the good things 5e introduced. Let's not go steps backwards, and instead focus on making things that are actually good.

It does mean casters have very little reason to lean into weapon use and need huge bonuses there to make it worth using even on subclasses focused around weapon use (see: Valor Bard [fail], Swords Bard, Bladesinger, Hexblade). It's kinda silly that weapons don't scale while cantrips do IMHO; I think weapons could afford to get extra dice while at it and then Extra Attack could multiply that so warriors could veritably be significant single target damage dealers.

[NaughtyTiger]

Level 5-10... They start acquiring more permanent magic items (uncommon and rare ones) as well, which will serve them for the rest of their careers.

The advice is that the DM starts to give out uncommon items to the party in Tier 2 per random table rolls. It does not say the martial should have a magical weapon by level 5 or such.

Advice from Xanathar's guide (published 3 years later) seems to recognize that martials are a bit hosed by a scarcity of magic weapons.

[PhoenixPhyre]

The advice is that the DM starts to give out uncommon items to the party in Tier 2 per random table rolls. It does not say the martial should have a magical weapon by level 5 or such.

Advice from Xanathar's guide (published 3 years later) seems to recognize that martials are a bit hosed by a scarcity of magic weapons.

You have a substantial probability of getting uncommon items even in T1. And remember, non-magical BPS resistant monsters aren't that common. Sure, if you're running a fiend (demon, not devil, because those can mostly be bypassed with silver) heavy campaign, you probably should give out more magic weapons. But if you're doing that, you're hosing the casters as well unless they can do force damage. Because fiends are resistant or outright immune to a lot of stuff.

But if you are, like most campaigns, doing heavy humanoids, beasts, and monstrosities through most of T1 and T2...it basically doesn't come up. And the few that do (were creatures) can be bypassed with silver.

Constructs? Adamantine (which is not magical and costs 500 gp to silver's 100gp).

So for this to be a problem, you need all of
1) no casters willing to use any spell slots to buff the martials. Which is a failure to play as a team unless you outright have no casters with any of holy weapon or magic weapon on their lists. In which case everyone's in the same boat.
2) no magic weapons, not even common ones (the Moontouched Blade bypasses resistance/immunity just fine and is common) or ammunition
3) facing enemies that don't have bypass methods such as silver and adamantine. Which are rather few and far between at low levels.

Really, the whole "need a magic weapon or I'm gimped" thing is way overblown.

[NaughtyTiger]

not even common ones (the Moontouched Blade bypasses resistance/immunity just fine and is common) or ammunition


you keep referencing Moontouched Blade. That was Xanthar's, right? 2017.
Common magical items weren't a thing in DMG.

WoTC seems to have acknowledged that the original design intent of magical weapons start near T2 hurt martials.

[PhoenixPhyre]

you keep referencing Moontouched Blade. That was Xanthar's, right? 2017.
Common magical items weren't a thing in DMG.

WoTC seems to have acknowledged that the original design intent of magical weapons start near T2 hurt martials.

No. Not really. Because...the number of resistant/immune creatures you'd face in T1 is basically zero.

There are the following number of creatures who are resistant (no immune creatures in the list) to non-magical attacks without a workaround (ie silver or adamantine) in CR 0-4:
2 CR 1/2 creatures (magmin, shadow)
5 CR 1 creatures (including the imp, who dies in 2 hits anyway even with resistance because its hit points suck)
5 CR 2 creatures (4 from the MM)
8 CR 3 creatures (one of whom is vulnerable to fire of all things, and only 3 of which are MM creatures)
7 CR 4 creatures, of which 3 (banshee, ghost, and couatl) are MM. Of which one is actually expected to really be a combat encounter (the banshee).
Total: 27 CR 0 - 4 in MM, VGtM, and MToF, of which 16 are MM.
And that's all. For reference, there are

32 CR 0
28 CR 1/8
59 CR 1/4
45 CR 1/2
55 CR 1
81 CR 2
56 CR 3
36 CR 4
Total 392 CR 0-4 creatures in those same books.

That's not even 7% of the total that are resistant (no immune). And there is exactly
0 humanoid
1 ooze (slithering tracker, VGtM)
2 monstrosities (grick [MM] and shadow mastiff [VGtM])
1 aberration (intellect devourer, generally not fought in T1 anyway)
1 construct (scarecrow, vulnerable to fire)
0 giants
0 beasts
0 fey

in that list. All of the rest are fiends, undead, or elementals. Specifically:
10 undead (6 from the MM, all but the Mummy are incorporeal and have crap tons of other immunities and resistances to energy types)
8 fiends (4 from the MM, two of which have crap for HP and no effective attacks)
3 elementals (2 from the MM, the fire snake (which is immune to fire) and the water weird, which is resistant to fire and immune to poison)

So no. I'd say "less than 7%, and basically 0 of the major types you're expected to encounter frequently at those levels" isn't any kind of serious issue. And, note...T1 cantrips are generally worse than a light crossbow even if the target is resistant to the piercing damage. T1 cantrips stink. And the riders don't mean anything at those levels (not that they mean much at any level).

[Witty Username]

So, cantrip damage, keep in mind a Tier 1 caster should NEVER, cast a damage cantrip for the goal of dealing damage outside of the specific case of warlock with agonizing blast. All casters have the option of a light crossbow instead, 1d8+dex will outdamage even a d12 cantrip unless there is a significant gap between dex and the casting stat. Without a potent cantrip style boost this holds true more or less until the Second damage increase at 11th level (tier 2 it is touch and go with about 1 point or so one way or the other, depending on the dex needs of the caster).

A fighter can deal 1d8+str or dex, twice at 5th level, without a fighting style and has more options to increase damage besides.

The idea that cantrips keep up with martials is nonsense.

[Talamare]

So, cantrip damage, keep in mind a Tier 1 caster should NEVER, cast a damage cantrip for the goal of dealing damage outside of the specific case of warlock with agonizing blast. All casters have the option of a light crossbow instead, 1d8+dex will outdamage even a d12 cantrip unless there is a significant gap between dex and the casting stat. Without a potent cantrip style boost this holds true more or less until the Second damage increase at 11th level (tier 2 it is touch and go with about 1 point or so one way or the other, depending on the dex needs of the caster).

A fighter can deal 1d8+str or dex, twice at 5th level, without a fighting style and has more options to increase damage besides.

The idea that cantrips keep up with martials is nonsense.
Casters generally have 14 Dex

1d8+2 = 6.5
1d12 = 6.5

So, do you want to gamble on your +4 to hit or your DC13 check

I'm pretty sure DC13 is a higher accuracy rate than a +4

[PhoenixPhyre]

Casters generally have 14 Dex

1d8+2 = 6.5
1d12 = 6.5

So, do you want to gamble on your +4 to hit or your DC13 check

I'm pretty sure DC13 is a higher accuracy rate than a +4

It isn't, actually. Save cantrips (which are dominantly lower damage already) are substantially less accurate than attacks. And cannot crit. A +4 in t1 is a 60% hit rate (across most foes). A DC 13 dex save (wisdom is worse) is about a 50% hit rate against actual monsters of the same CR.

So unless you can pick and choose (which with 1 or 2 damage cantrips you really can't) and always target the weak save (which many monsters at these levels don't have, really, as their con, Dex, and wis are all within a point or so), save cantrips are quite a bit worse than attack cantrips, which are worse than weapon attacks.

[Witty Username]

Casters generally have 14 Dex

1d8+2 = 6.5
1d12 = 6.5

So, do you want to gamble on your +4 to hit or your DC13 check

I'm pretty sure DC13 is a higher accuracy rate than a +4

Most of the time for me, casters have 16 dex to have their proper AC in poor armor, so 1d8+3 and +5 to hit for 7.5. And to be fair d12 is toll of the dead(not guaranteed as d12 is a damaged enemy) or poison spray(range 5ft) and going to be more commonly d8 damage from things like ray of frost, chill touch or sacred flame.
As for DC, it is often a wash given the variety of monsters and NPCs, and things like magic resistance crop up from time to time which will wipe the effectiveness of save effects. Also, it is easier to generate advantage on attack rolls generaly than disadvantage on saving throws. Unless you have one of the couple effects that allows half damage on cantrips (evocation wizard and I think one other thing I can't recall) DC isn't much of a difference.

[Talamare]

It isn't, actually. Save cantrips (which are dominantly lower damage already) are substantially less accurate than attacks. And cannot crit. A +4 in t1 is a 60% hit rate (across most foes). A DC 13 dex save (wisdom is worse) is about a 50% hit rate against actual monsters of the same CR.

So unless you can pick and choose (which with 1 or 2 damage cantrips you really can't) and always target the weak save (which many monsters at these levels don't have, really, as their con, Dex, and wis are all within a point or so), save cantrips are quite a bit worse than attack cantrips, which are worse than weapon attacks.

Most low CR monsters have awful Wis or Int
At DC13, you can expect a solid 60% accuracy rate, since most CR4 or lower monsters are at +0 to Int/Wis/Cha; or even negative

Also, you have the option of stuff like Firebolt; that trade essentially the +2 Damage for +1 Accuracy

Note, my stance isn't that they are better or good.
My stance is they are perfectly fine. "Eh, but it's Magic!"

[LibraryOgre]

I also cant help but notice that you accounted for potential misses with the fighter using GWM, but not potential saves made for Word of Radiance.

In general, AC and saving throws are relatively close in numbers, both at the beginning (where you have weaker equipment, but proficiency matters less), and later (where your equipment is better, and proficiency matters more). With bounded accuracy, they wind up largely on the same scale... your saving throws generally top out at +11 (+5 from stat, +6 from proficiency), and armor isn't too far different (18 for plate + shield puts you at 20, as does unarmored defense for monks and barbarians, while a really high dex light armor is about 16). So, given proficient saves, a target is about as likely to save against your cantrip as you are to miss with your attack. It results in mostly a wash in terms of hit probabilities.

GWM changes this, because it changes the hit probability radically. While it's a potentially big boost to damage, you're taking a 25% reduction in hit probability, which is a significant factor on the value of that bonus to damage; you're completely negating proficiency bonus for a good portion of the game. In many ways, save-based cantrips have the reverse; the roughly 66% chance that someone won't be proficient in a save makes it far more likely that you will hit. Target a fighter with Sacred Flame, and he's probably not going to be proficient in the save. At level 17, you've got a +11 to hit, and, at best, he has a +5 to save if non-proficient. He might have a -1 to save, against your DC of 19, if he dumped Dex because he was planning on wearing heavy armor.

[Talamare]

and armor isn't too far different (18 for plate + shield puts you at 20, as does unarmored defense for monks and barbarians, while a really high dex light armor is about 16).

I find AC in 5e a little hilarious...
Light Armor is 17
Medium is 17, 18 with Feat
Heavy is 18
All of which can use a Shield for 19~20
Monk is 20
Barbarian is 20... and can use a Shield for 22

Barbarian truly the Best Tank

[PhoenixPhyre]

Most low CR monsters have awful Wis or Int
At DC13, you can expect a solid 60% accuracy rate, since most CR4 or lower monsters are at +0 to Int/Wis/Cha; or even negative

Also, you have the option of stuff like Firebolt; that trade essentially the +2 Damage for +1 Accuracy

Note, my stance isn't that they are better or good.
My stance is they are perfectly fine. "Eh, but it's Magic!"

Not true. Because saves are already 10% worse than attacks due to the "who wins ties" difference. At +0 save/attack modifier, AC 10 is hit 55% of the time (hits on 10+), while a DC 10 save is failed (the success condition for the attacker) 45% of the time. And the save can't crit, which adds some (not much, but some) damage. Especially in later tiers with more dice.

And the number of monsters with better WIS defenses than AC (by % success) is pretty low. Looking at T1, Wis Saves % success (same modifiers in attack stats, same monsters) are 5% to 10% lower than attacks. And since you don't always get the juicy d12...even toll the dead looks pretty pitiful compared to others.

But I agree that cantrips are, roughly, fine where they are. They're no one's choice of weapons if they have other abilities, their riders are anywhere between null and small, and the scaling is fine. They're a rounding error in terms of balance. And if someone decides to use cantrips in T1--that's great. NBD. Because, again, rounding error.

But if you make them non-scaling, then they're
a) at best meh in T1
b) horrible and an active trap (worse than just about any other action the majority of the time) in T2+.

[Talamare]

Indeed, doesn't hurt much to have scaling cantrips

Eldritch Blast is being removed from the cantrip list in the nextDnD

So, everything seems fine to me

[PhoenixPhyre]

Indeed, doesn't hurt much to have scaling cantrips

Eldritch Blast is being removed from the cantrip list in the nextDnD

So, everything seems fine to me

Honestly, even EB + AB isn't great for a 2-level dip. It's a nice prize for someone who was already going to dip that far and needed a reliable ranged attack. But it's outshone by any dedicated martial weapon attack or real caster damage option using spell slots.

<unpopular>
Personally, I think they should fix multiclassing so dips aren't so attractive and leave EB alone.
</unpopular>

[LibraryOgre]

<unpopular>
Personally, I think they should fix multiclassing so dips aren't so attractive and leave EB alone.
</unpopular>

I'm a firm believer in "get rid of multiclassing, any multiclass should be your subclass, instead."

You're not a fighter/wizard. You're an Eldritch Knight, or an Arcane Warrior, which are different flavors of the same thing.

[Captain Cap]

I'm a firm believer in "get rid of multiclassing, any multiclass should be your subclass, instead."

You're not a fighter/wizard. You're an Eldritch Knight, or an Arcane Warrior, which are different flavors of the same thing.
That's basically the route taken by Pathfinder 2E.

[animorte]

I'm a firm believer in "get rid of multiclassing, any multiclass should be your subclass, instead.”

I have always felt the same. 5e did a fairly good job of creating the proper environment for that. My only problem with it is how inconsistent the subclass itself comes online. The general public’s problem with it is that the changes, no matter what, will never please everybody. The power levels just won’t be enough.

[PhoenixPhyre]

I'm a firm believer in "get rid of multiclassing, any multiclass should be your subclass, instead."

You're not a fighter/wizard. You're an Eldritch Knight, or an Arcane Warrior, which are different flavors of the same thing.

Agreed. And if they need more subclasses, great.

[Segev]

The trouble with "your multiclass is your subclass" is that it never actually works very well. You always are just bad at both classes, unless the subclass is actually bespoke for the class, in which case it never lets you actually BE the other class so much as a slightly flavored hybrid. If you want, for instance, to play a druid/necromancer, it's impossible because "necromancer" is a subclass and, if you can only multiclass by taking "nature wizard" subclass or "arcane druid" subclass to be a druid/wizard, you can never be a druid/necromancer.

[Witty Username]

Multiclassing is also a useful tool to represent change, the back-alley assassin trading their services to learn magic.
Arcane Trickster is a thought but that is more if you started as one as opposed as arrived at it through play.
I think you could get it with a modification of backgrounds as well as subclasses and rules for class changing, but it would require some tinkering with the rules a bit to make work.

[TaiLiu]

When the designers explicitly say that the game can be played just fine with no magic items at all, and that's not actually true, that's 100% the fault of the designers.
I guess I mostly agree. The game plays fine in the sense that it's perfectly playable. I just think the consequences of BPS-resistant and BPS-immune monsters mean that martials need a magic weapon or they get shafted when fighting those monsters. Which sucks, and which sucks needlessly. The writers could've made different design decisions.


The advice is that the DM starts to give out uncommon items to the party in Tier 2 per random table rolls. It does not say the martial should have a magical weapon by level 5 or such.

Advice from Xanathar's guide (published 3 years later) seems to recognize that martials are a bit hosed by a scarcity of magic weapons.
Right!


No. Not really. Because...the number of resistant/immune creatures you'd face in T1 is basically zero.

There are the following number of creatures who are resistant (no immune creatures in the list) to non-magical attacks without a workaround (ie silver or adamantine) in CR 0-4:
2 CR 1/2 creatures (magmin, shadow)
5 CR 1 creatures (including the imp, who dies in 2 hits anyway even with resistance because its hit points suck)
5 CR 2 creatures (4 from the MM)
8 CR 3 creatures (one of whom is vulnerable to fire of all things, and only 3 of which are MM creatures)
7 CR 4 creatures, of which 3 (banshee, ghost, and couatl) are MM. Of which one is actually expected to really be a combat encounter (the banshee).
Total: 27 CR 0 - 4 in MM, VGtM, and MToF, of which 16 are MM.
And that's all. For reference, there are

32 CR 0
28 CR 1/8
59 CR 1/4
45 CR 1/2
55 CR 1
81 CR 2
56 CR 3
36 CR 4
Total 392 CR 0-4 creatures in those same books.

That's not even 7% of the total that are resistant (no immune). And there is exactly
0 humanoid
1 ooze (slithering tracker, VGtM)
2 monstrosities (grick [MM] and shadow mastiff [VGtM])
1 aberration (intellect devourer, generally not fought in T1 anyway)
1 construct (scarecrow, vulnerable to fire)
0 giants
0 beasts
0 fey

in that list. All of the rest are fiends, undead, or elementals. Specifically:
10 undead (6 from the MM, all but the Mummy are incorporeal and have crap tons of other immunities and resistances to energy types)
8 fiends (4 from the MM, two of which have crap for HP and no effective attacks)
3 elementals (2 from the MM, the fire snake (which is immune to fire) and the water weird, which is resistant to fire and immune to poison)

So no. I'd say "less than 7%, and basically 0 of the major types you're expected to encounter frequently at those levels" isn't any kind of serious issue.
I think this list is right, and I appreciate the effort you took to put it together, but I think excluding workarounds is a mistake. I think the lycanthropes are the biggest problem, which are classic and cool and which bizarrely are vulnerable (not the game term) to all kinds of things that aren't silvered weapons. Those workarounds should be one of the few things that do in fact harm them.


And, note...T1 cantrips are generally worse than a light crossbow even if the target is resistant to the piercing damage. T1 cantrips stink. And the riders don't mean anything at those levels (not that they mean much at any level).
In T1, I'm actually really into vicious mockery and mind sliver, both of which have really cool riders. Create bonfire can be fun for campaigns that take place in dungeons, too.


The trouble with "your multiclass is your subclass" is that it never actually works very well. You always are just bad at both classes, unless the subclass is actually bespoke for the class, in which case it never lets you actually BE the other class so much as a slightly flavored hybrid. If you want, for instance, to play a druid/necromancer, it's impossible because "necromancer" is a subclass and, if you can only multiclass by taking "nature wizard" subclass or "arcane druid" subclass to be a druid/wizard, you can never be a druid/necromancer.
Isn't the Spores Druid basically the necromancer druid?

[Witty Username]

Isn't the Spores Druid basically the necromancer druid?

Sotra, in the same way evoker wizard is basically a dragon sorcerer wizard.

If necromancer is, able to cast animate dead, sure. I would expect at least a paring with a background like sage to carry the arcane knowledge angle as well.

[Segev]

Isn't the Spores Druid basically the necromancer druid?

Perhaps. Where is the Sword Bard Druid, the Spores Druid Bard, or the Fathomless Warlock Fighter?

[MrStabby]

I think that cantrips are not too strong.

For all the issues with some casters, it really isn't the cantrips that are an issue. Honestly, in most games it isn't even the average power of casters that is an issue it is their peak power - the ability to take away parts of the game from other players by "solving" them before others get to do their cool thing. Cantrips most definately don't contribute to this.

Cantrips are also a bit of a cruch if you are building themati characters rather than optimised characters. If you are a focusse enchanter wizard facing some encounters with charm immune enemies then you fall back on your cantrips, where a more optimised character would fall back on their other spells known. Cantrips as a back-up become more and more important the narrower your character focus. Cutting cantrips pushes players towards bland builds using the same spread of spells because they need to cover everything with their spell selecion, as the consequences of no doing so is being utterly ineffective.

It also comes down to how a character feels - if I play a cleric I want to feel like I am playing a cleric, which means doing cleric type things most turns and being happy with it. Likewise with a sorcerer or bard... If the background use of cleric powers in underwhelming, then I am not going to be happy (say if I am playing a healer and waiting for someone to be injured and alreay concentrating on bless or whatever). Making casters have crap turns isn't really a solution for them making other characters have crap turns under other circumstances... its just more crap turns in your game not fewer.

[sithlordnergal]

I'm a firm believer in "get rid of multiclassing, any multiclass should be your subclass, instead."

You're not a fighter/wizard. You're an Eldritch Knight, or an Arcane Warrior, which are different flavors of the same thing.

The problem is Eldritch Knights are not Arcane Warriors. They're not even close to that, because they don't have the spell slots to actually be "Arcane". They're a Fighter that dabbles in some magic. You do have Hexblade, Bladesinger, Valor Bard, and Sword Bard, but the only ones that feel like proper "Arcane Warriors" are Hexblades and Sword Bards. Bladesingers lean too heavily into the "Arcane" due to being a Wizard, while their "Warrior" aspect is limited to One-Handed Weapons without a Shield, and Light or No Armor. Valor Bards are less limited than Bladesingers when it comes to weapons and armor, but they only get one Extra Attack and lack anything to boost it like the others do, making it a fairly weak GISH class.

Now, this I will admit, the Bladesinger is perfect if you want a lithe, dexterous build. It excels at that sort of thing. But it is not so great if you want to, say, make a character that's similar to the Grey Knights, Isaac/Felix from Golden Sun, Eragon, Vader or Darth Malgus, and who knows how many mythological figures. Those types of characters typically carry around heavy weapons, wear heavy armor, and have plenty of magic to boot.

Of course, you could just play a pure Paladin, or you could go War Cleric. But while Paladins might make excellent Grey Knights, they're no Darth Vader or Eragon, characters that routinely use magic that is very much NOT Divine. War Clerics have a similar issue, on top of the fact that their Bonus Action attack is no replacement for Extra Attack.


No, the best way to get a proper Arcane Warrior would be via multiclass. A Fighter/Wizard is closer to a proper Arcane Warrior than any of the above subclasses, and you can tweak how much of each you want. Do you want to lean heavily into the Warrior aspect? Take more Fighter. Wanna lean more heavily into the Arcane? Take more Wizard. Wanna be balanced? Keep them close to the same level with each other. Sure, a pure Paladin can make a decent Grey Knight, but you need those Sorcerer levels to emulate things like firing a off a bolt of lightning and such.

And its not just Arcane Warriors that benefit from this. Wanna make a Fighter that specializes in Unarmed Combat? Well, the Unarmed Fighting Style certainly exists, but I feel like Fighter/Monks can do it better. Totem Barbarian and Moon Druids are fine on their own, but mix them together and you can really make something that feels like you're a rampaging, wild beast. The muscle of a criminal organization could be a Fighter or Rogue, but Fighter/Rogues and Barbarian/Rogues do it better.

Multiclassing just opens up so many options that it would be almost impossible to cover them all via subclasses alone.

[TaiLiu]

Sotra, in the same way evoker wizard is basically a dragon sorcerer wizard.

If necromancer is, able to cast animate dead, sure. I would expect at least a paring with a background like sage to carry the arcane knowledge angle as well.
Yeah, sage is good. Or even just proficiency in Arcana or Religion, regardless of background.

And yes, Spores Druids get animate dead. More frustrating are the necromantic warlock subclasses (and there are two!) that don't get access to that spell. Like... why?


Perhaps. Where is the Sword Bard Druid, the Spores Druid Bard, or the Fathomless Warlock Fighter?
I'm not totally sure I understand what you're talking about. Like, are you asking about thematics or, like, class mechanics? Cuz if the latter, well, you can't have it all (without multiclassing). That's the cost of a class-based system, and I think it's a reasonable one. Arguably this is even a feature.

[sithlordnergal]

I'm not totally sure I understand what you're talking about. Like, are you asking about thematics or, like, class mechanics? Cuz if the latter, well, you can't have it all (without multiclassing). That's the cost of a class-based system, and I think it's a reasonable one. Arguably this is even a feature.

While having limitations is a feature of a class based system, being able to multiclass naturally increases the number of options within that system. And its not like you get to have the stuff for free. While I won't deny classes are front loaded in 5e, you do have opportunity costs to multiclassing. Ranging from Ability Score requirements to losing levels, to potentially being less effective till late game. Its a give and take, and multiclassing tends to give a wider breadth of options.

[TaiLiu]

While having limitations is a feature of a class based system, being able to multiclass naturally increases the number of options within that system. And its not like you get to have the stuff for free. While I won't deny classes are front loaded in 5e, you do have opportunity costs to multiclassing. Ranging from Ability Score requirements to losing levels, to potentially being less effective till late game. Its a give and take, and multiclassing tends to give a wider breadth of options.
Yes, but that has problems. The writers didn't test for inter-class interactions. You gotta test for those, and if you're doing that you might as well switch to a points system instead. Multiclassing just combines many of the limits of a class-based system with many of the limits of a points system.

[sithlordnergal]

Yes, but that has problems. The writers didn't test for inter-class interactions. You gotta test for those, and if you're doing that you might as well switch to a points system instead. Multiclassing just combines many of the limits of a class-based system with many of the limits of a points system.

I mean, outside of the ability score thing, I actually think its a pretty decent system. I rather like it personally. I've never been a fan of point based systems personally, while they offer a lot of flexibility, I find their major weakness is being overly complicated for inexperienced players. I would say multiclassing combines the strengths and weaknesses of a class based system, while adding in enough flexibility to make the system interesting and expand it, without making it as complicated as a pure point based system.

Via multiclass, you really can make any sort of character work, can do so far better than a single class can, and without the complications of a point based system.

[LibraryOgre]

The problem is Eldritch Knights are not Arcane Warriors.

Which is why I called them different things.

The Eldritch Knight is an Eldritch Knight. Arcane Warrior? A fightery wizard subclass, instead of a wizardy fighter subclass. Might be bladesingers; I don't know them. But there's no reason you can't have a different fightery wizard subclass.

Hasn't been written that way, yet, of course.

I think one thing that comparing 3e and 5e multiclassing to 4e multiclassing is that the "take a level" version opens things up to odd synergies. The coffeelock. The Palock. Characters who excel because of the synergies of their abilities. 4e multiclassing, where it was a feat which gave you some limited abilities of the class? You didn't run into that, even if you wildly abused it (I played a 16th level bard with ONLY multiclassing feats. "Abuse" was kinda relative). TBH, there was even some of that back in AD&D, with human dual-classes... if you have the stats for it, toss a few thousand XP at another class, then go into your "real" class, especially if starting at higher level. Fighter was a popular start, but thieves made a great base for wizards (better HD and weapons, not intending on getting into a fight, anyway, and the 15 Dex was a great start on specialization).

Now, 5e has that 4e-like option built in. Want a splash of wizard? Magic Initiate. Want a splash of fighter? Lots of ways to make another character more fightery, from weapon and armor proficiencies to martial maneuvers.

[Witty Username]

Now, 5e has that 4e-like option built in. Want a splash of wizard? Magic Initiate. Want a splash of fighter? Lots of ways to make another character more fightery, from weapon and armor proficiencies to martial maneuvers.

Hm, I think that could work in concept, I think it runs into the issue that we need more classes for it to run smoothly, and we would need significantly more subclasses than we currently have.
And this is the second or third time I have thought that maybe we need to port the Duskblade from 3.5 to 5e, not sure why it keeps cropping up in my head.

[sithlordnergal]

Which is why I called them different things.

The Eldritch Knight is an Eldritch Knight. Arcane Warrior? A fightery wizard subclass, instead of a wizardy fighter subclass. Might be bladesingers; I don't know them. But there's no reason you can't have a different fightery wizard subclass.

Hasn't been written that way, yet, of course.
.

I feel like instead of making a subclass for it, you're just better off with the multiclass. They have taken shots at wizardy fighters via Valor Bard, Hexblade, Bladesinger, and Sword Bard, but none of them really can capture it the same way a Paladin/Sorcerer or Fighter/Wizard can. They certainly do their best, but they're still too limited.




I think one thing that comparing 3e and 5e multiclassing to 4e multiclassing is that the "take a level" version opens things up to odd synergies. The coffeelock. The Palock. Characters who excel because of the synergies of their abilities. 4e multiclassing, where it was a feat which gave you some limited abilities of the class? You didn't run into that, even if you wildly abused it (I played a 16th level bard with ONLY multiclassing feats. "Abuse" was kinda relative). TBH, there was even some of that back in AD&D, with human dual-classes... if you have the stats for it, toss a few thousand XP at another class, then go into your "real" class, especially if starting at higher level. Fighter was a popular start, but thieves made a great base for wizards (better HD and weapons, not intending on getting into a fight, anyway, and the 15 Dex was a great start on specialization).

Now, 5e has that 4e-like option built in. Want a splash of wizard? Magic Initiate. Want a splash of fighter? Lots of ways to make another character more fightery, from weapon and armor proficiencies to martial maneuvers.

I mean, I feel like 5e multiclassing is closer to 3e than 4e, and personally, I'm a big fan of 3.5's multiclassing, almost as much as I am of 5e's multiclassing. The synergy is part of what makes multiclassing fun. Heck, my favorite 3.5 build was a Ranger/Scout/Dervish because the classes had so much synergy. To the point where they even printed a feat in Complete Scoundrel called Swift Hunter that allowed you to treat your Scout levels as Ranger levels for Favored Enemy, and Ranger levels as Scout levels for Skirmish. It made for really fun and interesting characters that also had extra depth. I can't really say too much about 4e's multiclassing system because I know I'm biased already, I've never been a fan of 4e and basically continued to stick with 3.5 till 5e came out. That said, the Dual Classes of AD&D served a great purpose too.

And you're correct, 5e does have 4e-like options built into it, and if you only want a splash like that you're fine. But you need a lot more than just a few cantrips and a once-per-day spell in order to make something along the lines of a Grey Knight or Vader. You need actual magic, more than what Eldritch Knight gives you, along with actual Fighter abilities, more than what Bladesinger can offer. Hence the multiclass of Fighter/Wizard.

[TaiLiu]

I mean, outside of the ability score thing, I actually think its a pretty decent system. I rather like it personally. I've never been a fan of point based systems personally, while they offer a lot of flexibility, I find their major weakness is being overly complicated for inexperienced players. I would say multiclassing combines the strengths and weaknesses of a class based system, while adding in enough flexibility to make the system interesting and expand it, without making it as complicated as a pure point based system.

Via multiclass, you really can make any sort of character work, can do so far better than a single class can, and without the complications of a point based system.
You're right: it's easier to think about thirteen classes, their levels, and their subclasses than however many little skills and abilities a point-based system offers. But multiclassing has pitfalls that point-buy systems don't. Classes aren't just collections of mechanics, they're bound together by archetypes. That means that some intuitive multiclasses aren't great, and some unintuitive multiclasses are fantastic.

I was once in a campaign with a player who was a Rogue/Bard. This makes sense from a thematic point of view. They're both skill-focused classes, they're thematically similar, they seem like they would fit well. (Seems like it'd fit a swashbuckler-type character, a rogue with a lil bit of magic charm, the bard who's secretly a pickpocket.) But that's not really the case for mechanical reasons (unless you like to collect Expertises) and the most impressive thing that character did was successfully cast faerie fire at a crucial moment.

Meanwhile, Cleric 1 / Wizard 19 can be really fantastic. But it's an unintuitive multiclass and relies on mechanics that a new player may be surprised by. Inexperienced players may not be overwhelmed, but multiclassing well relies unexpectedly on mechanical mastery, which might be worse.

[yisopo]

Only a single mention for Mind Sliver in all the thread. Don’t you think it’s a very good cantrip?

[Segev]

I'm not totally sure I understand what you're talking about. Like, are you asking about thematics or, like, class mechanics? Cuz if the latter, well, you can't have it all (without multiclassing). That's the cost of a class-based system, and I think it's a reasonable one. Arguably this is even a feature.

"You can't have it all without multiclassing," is my point, yes. Multiclassing makes it possible to play the combination characters. Doing that ONLY through subclasses requires a proliferation of subclasses that is unheard-of outside of super-late edition, and likely would have already had multiclassing rules re-invented long prior. Even then, the staid nature of rigid subclasses would tend to make it less satisfying overall. "Dabbler" subclasses that have unique mechanics are good. Multiclassing that permits you to mix and match to build your concept is good. Both have their place, and one just can't replace the other.


Only a single mention for Mind Sliver in all the thread. Don’t you think it’s a very good cantrip?

It's an okay cantrip. You have to have save-or-suck spells ready to go right after it, though, or it's underwhelming.

[TaiLiu]

Only a single mention for Mind Sliver in all the thread. Don’t you think it’s a very good cantrip?
I like it, but you knew that already. It's good for teaming up with another spellcaster.


"You can't have it all without multiclassing," is my point, yes. Multiclassing makes it possible to play the combination characters. Doing that ONLY through subclasses requires a proliferation of subclasses that is unheard-of outside of super-late edition, and likely would have already had multiclassing rules re-invented long prior. Even then, the staid nature of rigid subclasses would tend to make it less satisfying overall. "Dabbler" subclasses that have unique mechanics are good. Multiclassing that permits you to mix and match to build your concept is good. Both have their place, and one just can't replace the other.
I agree that you can't just substitute one for another. But I think getting rid of multiclassing and developing a more robust system for combining class concepts is better.

Like, right now we have: subclasses (like Eldritch Knight), feats (like Ritual Caster), and multiclassing. All these different mechanical knobs lets you add bits of classes together. I'd prefer that we drop the latter and really clean up the former two. We preserve the best parts of a class-based system with the best parts of multiclassing. Not without its sacrifices, of course, but as I mentioned they're ones I'm willing to make.

[sithlordnergal]

I agree that you can't just substitute one for another. But I think getting rid of multiclassing and developing a more robust system for combining class concepts is better.

Like, right now we have: subclasses (like Eldritch Knight), feats (like Ritual Caster), and multiclassing. All these different mechanical knobs lets you add bits of classes together. I'd prefer that we drop the latter and really clean up the former two. We preserve the best parts of a class-based system with the best parts of multiclassing. Not without its sacrifices, of course, but as I mentioned they're ones I'm willing to make.

You might be willing, but multiclassing is why I continue to make characters. Single classed characters are far too boring to play. They don't have enough breadth, and are far too limiting and limited in their scope. I've only ever played two single classed characters, a Moon Druid and Arcana Cleric. Moon Druid was fun, Aecana Cleric was not.

You can refine the other two options as much as you like, but it's never going to be as good as multiclassing when it comes to combining classes. So may as well keep the multiclassing option. Now, you can still refine the above options, but no reason to strip multiclassing to what it was like in 4e.

[Keravath]

This is a great reason why cantrips shouldn't scale.
most damage cantrips have riders that offset the few points of damage from the ability modifier,

chill touch - healing/damage type
viscous mockery - disadvantage/damage type
thorn whip - auto-pull
acid splash/green flame blade - 2nd target
booming blade - mobility control
.... [saves instead of AC]

flexibility should reduce damage

In my opinion, flexibility has already reduced damage for these cantrips.

The baseline is firebolt or eldritch blast which by themselves do d10 damage and nothing else.
Then you have the d8 cantrips that usually have a modest but situationally useful rider effect like chill touch.
Then you have the d6 cantrips that do something else that can be more reliable or more commonly useful (eg thorn whip which can move the target)
Then you have the d4 cantrip (vicious mockery) which I don't really think is worth using most of the time.

Restricting these to scaling damage based on class level is just overkill in my opinion - might as well force multiclass casters to do nothing on their turns when they don't cast a spell since they won't have any scaling options to do damage except avoiding multiclassing and sticking with the one class so that the cantrips do scale.

Anyway, run the kind of game you like to play but if you want to nerf multiclassing spell casters you could just say that spell casters can't multiclass.

[NaughtyTiger]

f you want to nerf multiclassing spell casters you could just say that spell casters can't multiclass.
doesn't surprise me that folks assign imaginary arguments to me, then tell me they disagree.


In my opinion, flexibility has already reduced damage for these cantrips.
the flexibility lies in the number of choices available at an encounter, at an action

[LibraryOgre]

I mean, I feel like 5e multiclassing is closer to 3e than 4e, and personally, I'm a big fan of 3.5's multiclassing, almost as much as I am of 5e's multiclassing.

Which is the crux; I heartily dislike 3e/5e style multiclassing because of the synergy hacks. Because you wind up with things like coffeelocks or pallocks, and because it makes system mastery a much bigger part of effectiveness. A couple levels of Warlock is great for pretty much any paladin build because of the way warlocks spell slots scale. A couple levels of rogue can really liven up any monk build, because unarmed plus sneak attack plus cunning action makes you a lot more effective at what a monk does, arguably more effective than a pure monk at it, partially due to the damage boost from sneak attack, partially because cunning action frees up ki points.

While the scout/dervish/ranger build may have been fun and had great synergy, that goes back to my problem with it... you wind up with character creation as a losable mini-game, and losing that mini-game winds up with the main game being less fun as you're overshadowed by people who were better at the mini-game. Subclasses, even classes, will still have bad choices, but you don't need nearly the level of system mastery to play well.

[PhoenixPhyre]

Which is the crux; I heartily dislike 3e/5e style multiclassing because of the synergy hacks. Because you wind up with things like coffeelocks or pallocks, and because it makes system mastery a much bigger part of effectiveness. A couple levels of Warlock is great for pretty much any paladin build because of the way warlocks spell slots scale. A couple levels of rogue can really liven up any monk build, because unarmed plus sneak attack plus cunning action makes you a lot more effective at what a monk does, arguably more effective than a pure monk at it, partially due to the damage boost from sneak attack, partially because cunning action frees up ki points.

While the scout/dervish/ranger build may have been fun and had great synergy, that goes back to my problem with it... you wind up with character creation as a losable mini-game, and losing that mini-game winds up with the main game being less fun as you're overshadowed by people who were better at the mini-game. Subclasses, even classes, will still have bad choices, but you don't need nearly the level of system mastery to play well.

Very much agree with all of this.

Multiclassing having strong synergies means that either
1. the game is designed for you to have them, and not having them means you're sub-par
2. OR the game isn't designed for you to have them and now you're blowing the top off the curve by having them.

Either of which is bad. System-expected numbers should not be failable/missable. You shouldn't be able to trade them for other things or get extra of them. If multiclassing were to exist in some form, it should
1) be purely horizontal growth, never making you better at what you were already good at
2) come at a cost. Multiclassing should mean your prime thing doesn't progress (or progress as fast) as someone who didn't, and that effect should be meaningful. Deciding as a wizard to take a 1 level dip into cleric for heavy armor and martial weapons should cost more than "delay access to the new spell level (not slots, just spells) for one level".

[Witty Username]

I will say I prefer AD&Ds multiclassing and dual-classing to the 3+ system. For one, I very much like playing the character I want from level 1. 5 levels in fighter then wizard for the ideal break point feels really unsatisfactory for fighter/wizard. Subclasses do this better but you still have this weird get to level 3 to unlock your character, and the current fare like Eldritch Knight are handled better by 5 levels in fighter, than switch to wizard since you need to wait to unlock your character anyway (and you get some more options like Bard and Sorcerer if the int caster isn't your speed).

At level 1, I am playing fighter/mage, is very satisfying in comparison. I don't have this weird divide between my character concept and their abilities. And I don't need to build for now and when my character turns on.

Dual-classing has more utility in, "this character doesn't feel like this any more", I think the current multiclassing systems arose from a desire to make dual-classing more accessible. Which, led to alot of the problems above.

For subclasses, I think picking at level 1 would resolve alot of this for me. At first level, I'm an EK just feels alot smoother.

[Talamare]

Very much agree with all of this.

Multiclassing having strong synergies means that either
1. the game is designed for you to have them, and not having them means you're sub-par
2. OR the game isn't designed for you to have them and now you're blowing the top off the curve by having them.

Either of which is bad. System-expected numbers should not be failable/missable. You shouldn't be able to trade them for other things or get extra of them. If multiclassing were to exist in some form, it should
1) be purely horizontal growth, never making you better at what you were already good at
2) come at a cost. Multiclassing should mean your prime thing doesn't progress (or progress as fast) as someone who didn't, and that effect should be meaningful. Deciding as a wizard to take a 1 level dip into cleric for heavy armor and martial weapons should cost more than "delay access to the new spell level (not slots, just spells) for one level".

I've only been hearing and reading that PF2e does everything better than 5e

I would love to know how they handle MCing

[LibraryOgre]

2) come at a cost. Multiclassing should mean your prime thing doesn't progress (or progress as fast) as someone who didn't, and that effect should be meaningful. Deciding as a wizard to take a 1 level dip into cleric for heavy armor and martial weapons should cost more than "delay access to the new spell level (not slots, just spells) for one level".

Related to this, and another conversation about 5e: A cleric who dips wizard doesn't lose out on slots, but a fighter who dips ranger does lose out on his main deal, which is bonus attacks on the Attack action.


For subclasses, I think picking at level 1 would resolve alot of this for me. At first level, I'm an EK just feels alot smoother.

Which goes to 2e kits, or PF's archetypes. In WotR, I'm playing an Archaeologist Bard, who is essentially a bard/thief. But I'm also multiclassed to a Dragon Disciple, because that's how you play that game... because the numbers add up, meaning my asimaar bard is also part dragon, because being part dragon gives me benefits I want.

[MrStabby]

Which is the crux; I heartily dislike 3e/5e style multiclassing because of the synergy hacks. Because you wind up with things like coffeelocks or pallocks, and because it makes system mastery a much bigger part of effectiveness. A couple levels of Warlock is great for pretty much any paladin build because of the way warlocks spell slots scale. A couple levels of rogue can really liven up any monk build, because unarmed plus sneak attack plus cunning action makes you a lot more effective at what a monk does, arguably more effective than a pure monk at it, partially due to the damage boost from sneak attack, partially because cunning action frees up ki points.

While the scout/dervish/ranger build may have been fun and had great synergy, that goes back to my problem with it... you wind up with character creation as a losable mini-game, and losing that mini-game winds up with the main game being less fun as you're overshadowed by people who were better at the mini-game. Subclasses, even classes, will still have bad choices, but you don't need nearly the level of system mastery to play well.

I don't like the phrasing of 'losable mini-game' as it implies somewhat that people are playing to win.

Ideally I feel character creation should be about developing and realising a cool character concept and what tools are needed to make it work, rather than a constrained optimisation problem around what the maximum power is within a given class structure. Multiclassing plays to both but the power disparity between different elements means that you end up with characters that are each fine in their own way, but shouldn't be in a game together.

I guess I feel the same way about things like the Ravnica backgrounds and Eberron races that give additions to spell lists. They can be used to supercharge a character and eliminate weaknesses, or they can be used to develop cool characters that otherwise just wouldn't be viable.

I have no objections to system mastery being rewarded, but I feel it should come during play rather than character creation. Let it be something in the moment, not something you can look up in advance on the Internet before session zero.

[PhoenixPhyre]

I've only been hearing and reading that PF2e does everything better than 5e

I would love to know how they handle MCing

They don't as I understand it. Instead, there are <class> Devotion feats that give you some class feature. IIRC, they stack somewhat, so if you pick up Wizard Devotion a few times, you get better spellcasting. But it's still very limited.

It's basically 4e multiclass feats (not the Hybrid that came with PHB3).

Personally, I've been playing around with (another) parallel advancement mechanism to replace multiclassing. 4x during your leveling process (levels 3, 8, 13, and 18), you can either take an Emulation feature for a different class OR continue on with a Devotion feature from your own class.

Emulation features grant slightly scaling emulations of the other class's Core Feature. Those that grant Spellcasting do so in their own special, limited sort of way. For example, the Wizard Emulation track is entirely spellcasting. Each time you take it, you get better spell slots unless you already have spell slots from your class. You also count as a higher level in that class for things like spell preparation.



Times Taken
Wizard Cantrips
Minimum Spells in Spellbook
Slots (*)
Effective Wizard Level


1
3
3
2/0/0/...
1


2
3
6
4/3/0/...
4


3
4
9
4/3/3/1
7


4
4
12
4/3/3/2/1
10



So roughly half-progression. And you have to progress through the tracks linearly, although you can split up your Emulation between different classes (so taking 4 different classes each at 1 rank, 1 at 4 ranks, or 2 Devotion, 2 Emulation, etc).

Devotion features basically progress you on your Core Features faster, stronger, or otherwise help you be a little more <Class>-y. Things like extra infusions known. Extra spells prepared. Slightly better (non-moon druid) wildshape. Indomitable becoming actual Legendary Resistances if you take the Fighter Devotion 4 times. Bigger martial arts die (stacking with the normal progression). Things to give you incentives to stay in your main class, things that people who splash out would miss.

-----------

Effectively, I see a multi-tiered approach to "multiclassing" working best.

For dips (1-2 levels), use things like feats. They get a single, well-characterized feature emblematic of the class.
For splashes (3-6 levels), use subclasses.
For even splits, make new base classes.

In all of this, I personally believe that DMs should be less afraid of homebrew, especially tweaking something that exists to meet needs better.[1] Things like
* changing the schools for EK or even the spell list (making a Druidic Knight, for instance).
* making new subclasses or feats based on existing ones.
* even reworking base classes.

Since you're only really working with a single table at a time, balance can happen relative to the other player characters there and the campaign at hand and doesn't have to match some abstract holistic notion of balance vs all the other options. And you can just say "no, you can't combine it with X" or whatever, because you're at a single table with a single group of people.

[1] and more skeptical and choosy about official material, which is just homebrew in a pretty package. Often worse than stuff you find on various sites, to be honest.

[Talamare]

They don't as I understand it. Instead, there are <class> Devotion feats that give you some class feature. IIRC, they stack somewhat, so if you pick up Wizard Devotion a few times, you get better spellcasting. But it's still very limited.

It's basically 4e multiclass feats (not the Hybrid that came with PHB3).

Personally, I've been playing around with (another) parallel advancement mechanism to replace multiclassing. 4x during your leveling process (levels 3, 8, 13, and 18), you can either take an Emulation feature for a different class OR continue on with a Devotion feature from your own class.

Emulation features grant slightly scaling emulations of the other class's Core Feature. Those that grant Spellcasting do so in their own special, limited sort of way. For example, the Wizard Emulation track is entirely spellcasting. Each time you take it, you get better spell slots unless you already have spell slots from your class. You also count as a higher level in that class for things like spell preparation.



Times Taken
Wizard Cantrips
Minimum Spells in Spellbook
Slots (*)
Effective Wizard Level


1
3
3
2/0/0/...
1


2
3
6
4/3/0/...
4


3
4
9
4/3/3/1
7


4
4
12
4/3/3/2/1
10



So roughly half-progression. And you have to progress through the tracks linearly, although you can split up your Emulation between different classes (so taking 4 different classes each at 1 rank, 1 at 4 ranks, or 2 Devotion, 2 Emulation, etc).

Devotion features basically progress you on your Core Features faster, stronger, or otherwise help you be a little more <Class>-y. Things like extra infusions known. Extra spells prepared. Slightly better (non-moon druid) wildshape. Indomitable becoming actual Legendary Resistances if you take the Fighter Devotion 4 times. Bigger martial arts die (stacking with the normal progression). Things to give you incentives to stay in your main class, things that people who splash out would miss.

-----------

Effectively, I see a multi-tiered approach to "multiclassing" working best.

For dips (1-2 levels), use things like feats. They get a single, well-characterized feature emblematic of the class.
For splashes (3-6 levels), use subclasses.
For even splits, make new base classes.

In all of this, I personally believe that DMs should be less afraid of homebrew, especially tweaking something that exists to meet needs better.[1] Things like
* changing the schools for EK or even the spell list (making a Druidic Knight, for instance).
* making new subclasses or feats based on existing ones.
* even reworking base classes.

Since you're only really working with a single table at a time, balance can happen relative to the other player characters there and the campaign at hand and doesn't have to match some abstract holistic notion of balance vs all the other options. And you can just say "no, you can't combine it with X" or whatever, because you're at a single table with a single group of people.

[1] and more skeptical and choosy about official material, which is just homebrew in a pretty package. Often worse than stuff you find on various sites, to be honest.

Thank you for the information! I think it would have been a slight pain to read their Core just to try to find it.

[TaiLiu]

You might be willing, but multiclassing is why I continue to make characters. Single classed characters are far too boring to play. They don't have enough breadth, and are far too limiting and limited in their scope. I've only ever played two single classed characters, a Moon Druid and Arcana Cleric. Moon Druid was fun, Aecana Cleric was not.

You can refine the other two options as much as you like, but it's never going to be as good as multiclassing when it comes to combining classes. So may as well keep the multiclassing option. Now, you can still refine the above options, but no reason to strip multiclassing to what it was like in 4e.
Sure, that's reasonable. I doubt 5.5e will remove multiclassing. In 5e it's "optional," but I've never been in a campaign where you can't multiclass.

I guess we generally agree on the facts but have different viewpoints. I'd argue that the point of classes is to place characters into broad archetypes, so those limits are in fact a good thing. And limited flexibility can be gained with feats or whatever, and those limits are also good. Of course, D&D doesn't do that well compared with, say, Monster of the Week.


Which is the crux; I heartily dislike 3e/5e style multiclassing because of the synergy hacks. Because you wind up with things like coffeelocks or pallocks, and because it makes system mastery a much bigger part of effectiveness.
Good point. I think that's right, although I think the mechanical crunch of D&D is always gonna implicitly have "hacks" and reward system mastery. Even with multiclassing gone.


Related to this, and another conversation about 5e: A cleric who dips wizard doesn't lose out on slots, but a fighter who dips ranger does lose out on his main deal, which is bonus attacks on the Attack action.
Yeah, the way they did martial and spellcaster multiclassing in 5e isn't great.


have no objections to system mastery being rewarded, but I feel it should come during play rather than character creation. Let it be something in the moment, not something you can look up in advance on the Internet before session zero.
Right!

[LibraryOgre]

I don't like the phrasing of 'losable mini-game' as it implies somewhat that people are playing to win.

Some people really are; I had one at my last game who'd come up with a ridiculously powerful build, and had a couple other players on similarly powerful builds, each time taking advantage of synergies within the system.

And, in other cases, there's the chance to just *lose*. It's a lot harder in 5e v. 3.x, to be sure, but there was always the chance of realizing you needed specific stats to get necessary feats, or choosing the wrong feat, or a trap feat.



Ideally I feel character creation should be about developing and realising a cool character concept and what tools are needed to make it work, rather than a constrained optimisation problem around what the maximum power is within a given class structure. Multiclassing plays to both but the power disparity between different elements means that you end up with characters that are each fine in their own way, but shouldn't be in a game together.

Sure, it should be... but it isn't always, and take-a-level multiclassing increases the possibility for those synergies to cause problems, thus my distaste for it.

[Psyren]

Related to this, and another conversation about 5e: A cleric who dips wizard doesn't lose out on slots, but a fighter who dips ranger does lose out on his main deal, which is bonus attacks on the Attack action.

But that cleric does lose out on high-level spells, or at the very least delay them. Those are usually much more valuable than the slots. Heal >>> upcast Cure Wounds for instance.

The continued slot progression is closer to being a consolation prize than a balance issue.

[PhoenixPhyre]

But that cleric does lose out on high-level spells, or at the very least delay them. Those are usually much more valuable than the slots. Heal >>> upcast Cure Wounds for instance.

The continued slot progression is closer to being a consolation prize than a balance issue.

Two things--
* cleric's higher-level spells are...not much to write home about.
* Wizards are, but the value of a single level dip the other way (ie wizard X/Cleric 1) for armor and some low-level cleric spells that, while yes, not equal to the high-level replacements, still upcast pretty darn well, is tremendous.

[Psyren]

Two things--
* cleric's higher-level spells are...not much to write home about.
* Wizards are, but the value of a single level dip the other way (ie wizard X/Cleric 1) for armor and some low-level cleric spells that, while yes, not equal to the high-level replacements, still upcast pretty darn well, is tremendous.

I don't disagree (with the second part anyway), but at the end of the day multiclassing is a privilege, and the system did its job by clearly labeling it as such. If you need to ban or modify it to fit your desired power level your players need to be okay with that. I'd still much rather have it as an option than not, even in its current form.

The armor thing in particular can be fixed with a tiny tweak. It can be absolute (you can't cast wizard spells in any kind of physical armor, period) or softened (you can go one step lighter than the strongest category you have proficiency in, and you need War Caster to use shields), etc.

That won't help you with DDAL of course - but then, I suspect that just as you wouldn't want to run into AL players who favor multiclassing and dips there, they probably wouldn't be keen on running into DMs who hate it either.

[PhoenixPhyre]

I don't disagree (with the second part anyway), but at the end of the day multiclassing is a privilege, and the system did its job by clearly labeling it as such. If you need to ban or modify it to fit your desired power level your players need to be okay with that. I'd still much rather have it as an option than not, even in its current form.

The armor thing in particular can be fixed with a tiny tweak. It can be absolute (you can't cast wizard spells in any kind of physical armor, period) or softened (you can go one step lighter than the strongest category you have proficiency in, and you need War Caster to use shields), etc.

That won't help you with DDAL of course - but then, I suspect that just as you wouldn't want to run into AL players who favor multiclassing and dips there, they probably wouldn't be keen on running into DMs who hate it either.

Multiclassing is a bad design, slapped on without any thought to keep 3e people happy. It is the source of a large fraction of all the issues with balance. It should get reworked to actually work instead of causing problems. It's a hack. And I have fundamental aesthetic issues with hack jobs. And adding more hacks on top offends me from a design perspective.

Multiclassing is the Javascript of D&D. Never designed for what it ended up being used for, and mostly just a vector of bad things.

[Segev]

If you want to say it was badly designed, that's one thing. Saying multiclassing is "bad design" just as a blanket statement is, I think, incorrect.

[Psyren]

Multiclassing is a bad design, slapped on without any thought to keep 3e people happy. It is the source of a large fraction of all the issues with balance. It should get reworked to actually work instead of causing problems. It's a hack. And I have fundamental aesthetic issues with hack jobs. And adding more hacks on top offends me from a design perspective.

Multiclassing is the Javascript of D&D. Never designed for what it ended up being used for, and mostly just a vector of bad things.

I'm not saying it's perfect. But there's a baby careening out of that bathwater.

[Keltest]

I'm not saying it's perfect. But there's a baby careening out of that bathwater.

Agreed. Most of the problems with multiclassing imo come from certain (sub)classes having a lot of game changing mechanics front loaded relative to other classes. Looking at you hexblade.

[Person_Man]

And, in other cases, there's the chance to just *lose*. It's a lot harder in 5e v. 3.x, to be sure, but there was always the chance of realizing you needed specific stats to get necessary feats, or choosing the wrong feat, or a trap feat.

On a related note, I think its fairly common for new players to pick the “wrong” ability scores because “it makes sense for my character” even though it mechanically sucks. And then they can get frustrated when it doesn’t work well.

5e is way better about avoiding this than 1/2/3E. (Though 4E was very balanced but had a legion of other issues). But there are a lot of issues with ability scores which could easily fixed. Each ability score should provide some real benefit irrespective of class choice, odd ability scores shouldn’t be a meaningless waste, every character should have the same total starting bonuses even if generated randomly, doubling up high Str and Dex shouldn’t be a complete waste, etc.

[PhoenixPhyre]

I'm not saying it's perfect. But there's a baby careening out of that bathwater.


Agreed. Most of the problems with multiclassing imo come from certain (sub)classes having a lot of game changing mechanics front loaded relative to other classes. Looking at you hexblade.

To mutate the metaphor, I don't believe there's a bathtub. There's certainly a baby, but what you call a bathtub, I call a krakenoid monster trying to devour the baby.

If the cost of keeping multiclassing around is that classes have to be more backloaded so you can't actually play them effectively starting by 3rd level (I'd prefer having all core features by level 1 or 2)...then that's not a cost worth keeping in my mind. This is warping one part of the game (the preferred path, since multiclassing is a variant tacked on without playtest) to keep the cancer at bay. And yes, I do believe multiclassing is cancerous. Introducing it inevitably means that it takes over and the game shifts until it's the favored (and only viable, at many tables) means. Which, as I've said before, is doing point-buy except extremely badly.

There are many ways of keeping the good parts (ability to build varied characters) without the distortions caused by level-by-level multiclassing. Which, may I remind you, has only really existed in one edition (3e) that it then proceeded to eat alive. Every other edition has looked at it and decided "yeah, that's not worth it." And I'm not aware of any other class-based game that has gone down this route other than PF1e, which is basically 3.75e and had it baked into the cake.

[Segev]

If you want to try to implement 1e and 2e style multiclassing (rather than the dual-classing that 3e-style multiclassing actually is), you'd need to either go back to XP calculation or you'd need to come up with a good way to balance advancing multiple classes simultaneously. But you could do it.

Easiest would be to use XP totals, and divide XP evenly between all of your classes.

[sithlordnergal]

Some people really are; I had one at my last game who'd come up with a ridiculously powerful build, and had a couple other players on similarly powerful builds, each time taking advantage of synergies within the system.


I mean, you're always going to have optimizers, even in a game with very little synergy between different classes/builds. Whenever you have two things that are comparable, people will compare them and find the most optimized option. Now, I do agree that it can be difficult to run for a hyper-optimized build if its the only optimized build in the party. I recently had that happen with an Artificer. They were so much stronger than anyone else in the party at T2, it was ridiculous. But I had a chat with the player, they agreed to swap to a new character, problem was solved.

That said, it only becomes an issue if they're the only optimized build in the party. If at least half of the party is optimized, hyper optimization isn't an issue cause you can just scale the encounters to the optimizers. If one player is falling behind, offer to help improve their build without messing with the base character traits.

And I don't think its "taking advantage of synergies within the system", its literally just optimization.



And, in other cases, there's the chance to just *lose*. It's a lot harder in 5e v. 3.x, to be sure, but there was always the chance of realizing you needed specific stats to get necessary feats, or choosing the wrong feat, or a trap feat.


I'll be honest...while it was very easy to "lose" character creation in 3.x, that's not really the case in 5e. There really aren't any trap options, especially now with the splat books. Its really, really hard to make a bad character in 5e. I can only name three characters that were actually bad. One of them is Abserd from Puffin Forest, and one was specifically designed to be a weak character, and the last was from a player that tried to play a Lore Bard as a frontline Fighter and preferred making melee attacks over casting spells.




Sure, it should be... but it isn't always, and take-a-level multiclassing increases the possibility for those synergies to cause problems, thus my distaste for it.

Unless you completely remove the ability for other classes to meaningfully interact with each other at all, there's always going to be optimization. And there are always going to be optimizers that are only interested in optimizing. Even if you only had feats like Magic Initiate and the like, you'd see optimizers finding synergies between feats and classes in order to optimize.

And I don't actually think optimizing on those synergies actually cause that many problems. You can get a stronger character, yes, but they're generally not that much stronger than the rest of the party. I won't deny that a Fighter/Wizard is strong, but I've never had any issues with one being "too strong". Heck, the Artificer that was too strong? Single classed character, no multiclassing at all. They simply took the most powerful infusions with a fire resistance race, and their power spiked well above what they normally should be capable of.

[sithlordnergal]

To mutate the metaphor, I don't believe there's a bathtub. There's certainly a baby, but what you call a bathtub, I call a krakenoid monster trying to devour the baby.

If the cost of keeping multiclassing around is that classes have to be more backloaded so you can't actually play them effectively starting by 3rd level (I'd prefer having all core features by level 1 or 2)...then that's not a cost worth keeping in my mind. This is warping one part of the game (the preferred path, since multiclassing is a variant tacked on without playtest) to keep the cancer at bay. And yes, I do believe multiclassing is cancerous. Introducing it inevitably means that it takes over and the game shifts until it's the favored (and only viable, at many tables) means. Which, as I've said before, is doing point-buy except extremely badly.

There are many ways of keeping the good parts (ability to build varied characters) without the distortions caused by level-by-level multiclassing. Which, may I remind you, has only really existed in one edition (3e) that it then proceeded to eat alive. Every other edition has looked at it and decided "yeah, that's not worth it." And I'm not aware of any other class-based game that has gone down this route other than PF1e, which is basically 3.75e and had it baked into the cake.

I fully disagree with this. The so-called "preferred path" is actually extremely boring. You don't actually have any interesting decisions to make once you decide your Race and Class. The only interesting decisions you make after taking your class and subclass are feats versus ASIs...and those aren't very interesting because both are an extremely limited resource. Its fine for new players to learn the system, but its very boring over all.

Multiclassing is what helps keep characters varied and the system interesting. If you're going for a multiclass build you need to decide which abilities are going to be put off till later, and which are worth losing entirely. You have to decide how/when you'll multiclass depending on the level you expect to reach, and there are exponentially more combinations then what the Class/Subclass system offers on its own. It also allows for stronger ties between mechanics and narrative. An Ancients Paladin/Druid feels far more like a a Warrior of Nature then either of those classes on their own, and far more than a Nature Cleric does. And the mechanics of those two classes uphold that feeling.

As for classes having all their abilities at level 2, I actually think its better for them to come online at level 3. Not because of multiclassing mind you, but to better ease in new players. Levels 1 and 2 aren't for experienced players, they're to help new players learn the system before they have a bunch of information and abilities dumped onto them via Subclasses. As someone who does run for new players, and I mean really new players that don't even own a copy of the PHB, they would be completely overwhelmed if they got all of their subclass features at level 2. Heck, one party was overwhelmed when they got their subclasses at level 3.

And finally, I think the fact that multiclassing is just a weaker form of point buy is a benefit. I find most point buy systems are confusing for newer players, and its very easy for a new player to make a bad character. They're new player hostile, because you're expected to know how the different abilities interact with each other before you've even made your first character. You're just given a bunch of tables, and you don't really learn how the different things interact. It might give suggestions on how to make a basic character...but very few people actually use the premade characters in the book.

Multiclassing allows for more flexibility without requiring the same level of system mastery that point buy does. Its not hostile to new players the way point buy is, because its not required for new players to use. New players will instead stumble upon them more naturally, as they learn and master the system on their own. They'll look through the book and realize "Hey, Class A and Class B look like they'd work well together" after having played Class A long enough to learn its strengths and weaknesses.

[animorte]

I fully disagree with this. The so-called "preferred path" is actually extremely boring. You don't actually have any interesting decisions to make once you decide your Race and Class. The only interesting decisions you make after taking your class and subclass are feats versus ASIs...and those aren't very interesting because both are an extremely limited resource. Its fine for new players to learn the system, but its very boring over all.

My turn to disagree with you but I’m only touching on this first part. The rest about multi-classing just looks like the exact type of conversation that people have differing opinions on, take it or leave it, go for it if you like it and ignore it if you don’t…

Blatantly a discussion that can only talk itself in circles. Some classes are stronger without it, other classes function well with the smallest dip, and a few are in desperate need of it. This is a bit more of a balance issue.

Anyway, onward to my point, and I’ll get right to it. You saying that it’s boring to pick a race and class then, nothing else except some feats/ASIs gives the impression that you aren’t particularly interested in actually playing, just building. I love building PCs too! But there are a lot more decisions to be made than what you present.

I can decide between battle maneuvers, invocations, infusions, how to use my Ki, tons of spells…
I can choose what my character is interested in, stuff about their past, their goals, how they interact with friends/neutral/enemies…

If you honestly feel there aren’t other decisions to be made, I just don’t even know where to begin.

Though I do understand where you’re coming from if every PC must be optimized in such a way that this class must always use one of these few races, must always choose the correct weapon, must always select the same spells… Sure, if optimizing is the only real goal here.

If so, RSP would like a word with you.
\/

[Psyren]

1) I agree that straight-classed builds can still be fun but the system would be so, so much more two-dimensional and dull without multiclassing. Especially back when the system launched and we didn't have even SCAG yet. Did they make some mistakes with the current rules, I'd say so (casting in full plate + shield as long as you're proficient with no further build investment being the biggest one), but they got some things right too, like not needing to track 2+ different classes worth of spell slots in two different places like you did in 3.5.

2) I really don't get the point-buy comment, there are numerous multiclass builds that don't have any functional difference in ASI choice than a straight-class one. Ranger/Monk builds want Dex/Wis just like their parent classes, Artificer/Wizard builds want Int/Con, Sorlock builds want Cha/Con etc. I don't see how multiclassing is any form of point buy for them.

[LibraryOgre]

And I don't think its "taking advantage of synergies within the system", its literally just optimization.

I mean, that's exactly what optimization is. And, while there will always be optimizers, the problem comes to the amount which optimization matters. In 2e, it was an optimizing choice, frequently, to take the long sword. Not only was it a good weapon, with which elves got a bonus, but it was the most likely weapon to be randomly generated as magical (most magical weapons were swords, and a plurality of swords were long swords). You might also optimize by starting as a fighter then dual-classing, getting the kicker of HP and weapon specialization.

However, when optimization puts a character on another plane compared to unoptimized characters, you make a game where system mastery can make or break the experience. As I've said, 5e is a lot better about this than 3e was, but you still have notable synergy problems that can arise. These are made more apparent when you cross classes.

I've never said that optimization is bad, in and of itself. I've never said that you won't have synergies in other systems or other ways. Rather, I've said that take-a-level multiclassing
a) increases the possibility for those synergies to exist
b) is elegantly handled by a combination of subclasses and feats.



If you want to try to implement 1e and 2e style multiclassing (rather than the dual-classing that 3e-style multiclassing actually is), you'd need to either go back to XP calculation or you'd need to come up with a good way to balance advancing multiple classes simultaneously. But you could do it.


Oh, gods, why would anyone want to do that? AD&D style multiclassing was horrible in so many ways; just an utter kludge of a system.

[PhoenixPhyre]

2) I really don't get the point-buy comment, there are numerous multiclass builds that don't have any functional difference in ASI choice than a straight-class one. Ranger/Monk builds want Dex/Wis just like their parent classes, Artificer/Wizard builds want Int/Con, Sorlock builds want Cha/Con etc. I don't see how multiclassing is any form of point buy for them.

Point buy as in "you have a pool of points with which you buy all your abilities" not the "how do you generate your ability scores" point buy. Point buy as in GURPS or HERO system or the Storyteller system, et al. Where buying "super strength" might have different points at different values, generally escalating as you buy more points.

The contrast is a class/level system, where you get a package, generally fixed (or "choose from these few fixed options") and then your growth happens dominantly within that class framework via large discrete chunks where many class-set values all go up at once.

Point-buy systems work when they can accurately cost various features. Multiclassing breaks that, because it assumes that the value of a level is always 1 level up. It's not, and it's not even transitive--taking your first level in Fighter and second in Wizard gives you major things that taking your first in wizard and second in fighter doesn't. The cost is not only not constant, it's not even decidable in the abstract. That makes it impossible to balance via standard point-buy mechanisms as well as throwing out the core of a class/level system by construction.

[Segev]

Oh, gods, why would anyone want to do that? AD&D style multiclassing was horrible in so many ways; just an utter kludge of a system.

It does address the complaints made about level-by-level multiclassing. You pick what you're going to do at chargen, and you pay for the increased versatility with much less power. You're not dipping anything for a shallow investment, either; you commit to the multiclass or you don't do it.

[sithlordnergal]

Anyway, onward to my point, and I’ll get right to it. You saying that it’s boring to pick a race and class then, nothing else except some feats/ASIs gives the impression that you aren’t particularly interested in actually playing, just building. I love building PCs too! But there are a lot more decisions to be made than what you present.

I can decide between battle maneuvers, invocations, infusions, how to use my Ki, tons of spells…
I can choose what my character is interested in, stuff about their past, their goals, how they interact with friends/neutral/enemies…

If you honestly feel there aren’t other decisions to be made, I just don’t even know where to begin.

Though I do understand where you’re coming from if every PC must be optimized in such a way that this class must always use one of these few races, must always choose the correct weapon, must always select the same spells… Sure, if optimizing is the only real goal here.

If so, RSP would like a word with you.
\/

So, those are choices, but most of those are a very different kind of choice. Not all choices are carry the same kind of weight and not all choices are really comparable Heck, I wouldn't even count things like goals, character interests, how they interact with NPCs, and encounter tactics to even be in the same category as choices made during character creation or leveling. They are choices, but so far from the kinds of choices I'm talking about that I wonder why one would bother bringing them up. Its like comparing apples to oranges.

Now, things like Infusions, Spells, Invocations, and Maneuvers are much closer to the choices I'm talking about. But the biggest issue is that they're class specific choices. As a result you'll either never make those choices, based on the class you chose, or you'll have to make the exact same choice. Sure, spell selection is important, but there's no real difference between deciding if you want Chaos Bolt or Catapult at level 1 and Meteor Swarm or Wish at level 20. Important choices to be sure, but very different from deciding if you want to give up your capstone, a 7th level slot, and a Sorcery Point in order to wear armor and use martial weapons, or giving up the previous stuff, an entire ASI, and another Sorcery Point for the ability to act twice on your turn once per short rest. There's a difference in kind between a choice like that and deciding what spell or invocation you learn.




However, when optimization puts a character on another plane compared to unoptimized characters, you make a game where system mastery can make or break the experience. As I've said, 5e is a lot better about this than 3e was, but you still have notable synergy problems that can arise. These are made more apparent when you cross classes.


See, I would agree with you if this were 3e. An optimized 3e character is leagues better than an unoptimized one. However, while that problem does exist in 5e, I don't think its caused by multiclassing. In fact, the classes I've found that cause the most issues when it comes to optimized vs unoptimized characters are almost always single classed. Classes like the Druid and Artificer are very much on a different plane compared unoptimized characters when they're optimized. A single Moon Druid can match or outperform the strength of an unoptimized party in Tier 1, and still be a major threat in T2 and above. Artificers...well...the only character I ever had to have a player swap out due to being too powerful was an Artificer in Tier 2, and I'm a pretty lenient DM.

Meanwhile, multiclassed characters have never given me an issue. They tend to have more breadth than single classed characters, but lose out on the power single classed characters gain. Sure, the Paladin/Sorcerer in my party gets to smite more often than the pure Paladin, and can rely on their magic more, but in exchange they go down much faster, their auras aren't nearly as good, and they can;t heal/support the party as effectively. If you asked me to genuinely pick which was the "stronger" or "better" character, I honestly couldn't tell you. Yeah, the Fighter/Wizard typically has a higher AC than the normal Wizard, but they're behind on spells, spell slots, they lose out in Int saves, which have become more common, and now they're a prime target for abilities that affect things being worn, like Rust Monsters or Heat Metal.




I've never said that optimization is bad, in and of itself. I've never said that you won't have synergies in other systems or other ways. Rather, I've said that take-a-level multiclassing
a) increases the possibility for those synergies to exist
b) is elegantly handled by a combination of subclasses and feats.


True, you've never said optimization is bad, but optimization is always on a scale. If we never had multiclassing, I'm sure the discussion would be about Feats that give you access to spells, Battlemaster Maneuvers and Fighting Styles, or Armor proficiency. And personally, one of the reasons I like level by level multiclassing so much is

a) I want to increase the possibility of those synergies to exist, I feel it makes the system better
b) Subclasses and feats do a horrific and half-baked job of trying to replicate multiclassing

You need multiclassing to get that system that subclasses and feats would try to, poorly, emulate.

[Psyren]

Point buy as in "you have a pool of points with which you buy all your abilities" not the "how do you generate your ability scores" point buy. Point buy as in GURPS or HERO system or the Storyteller system, et al. Where buying "super strength" might have different points at different values, generally escalating as you buy more points.

The contrast is a class/level system, where you get a package, generally fixed (or "choose from these few fixed options") and then your growth happens dominantly within that class framework via large discrete chunks where many class-set values all go up at once.

Point-buy systems work when they can accurately cost various features. Multiclassing breaks that, because it assumes that the value of a level is always 1 level up. It's not, and it's not even transitive--taking your first level in Fighter and second in Wizard gives you major things that taking your first in wizard and second in fighter doesn't. The cost is not only not constant, it's not even decidable in the abstract. That makes it impossible to balance via standard point-buy mechanisms as well as throwing out the core of a class/level system by construction.

Got it, that makes more sense. But I think bold is an assumption on your part that's never stated anywhere. The only way for class level A = class level B is if they're all the same, and that is the antithesis of depth. Even if that goal were somehow desirable, the amount of design time spent iterating to get there wouldn't be.

[Rukelnikov]

Got it, that makes more sense. But I think bold is an assumption on your part that's never stated anywhere. The only way for class level A = class level B is if they're all the same, and that is the antithesis of depth. Even if that goal were somehow desirable, the amount of design time spent iterating to get there wouldn't be.

His point is that given there is zero granularity in the "point" you use to buy, which is the level up, all classes at all levels must be equally balanced for the system to be balanced.

That's wrong though, because it doesn't happen in any PB system I've played (about 5 or 6 that come to mind rite nao). For instance in WoD, if I've a character built for combat and have low mental stats, taking my first dot in academics probably won't be as useful as for someone who is playing a very intelligent hacker and takes their first dot in the same skill. So same cost for both, different value for cost across different characters. In the same vein, if the combat oriented character took their first dot in stealth, they are likely to get more out of it than from academics, for the simple reason that stealth is usually rolled with Dex, and as a combatant will likely have a decent Dex, different value for the cost across different skills.

And that's the "old paradigm" in the early 10s edition cost are not increasingly higher, so taking your first dot in a skill or your 5th cost the same, the value you get from your 5th point in anything is almost assuredly gonna be much higher than taking the first dot in anything, for the simple reason that you are getting better at one of your main schticks instead of getting a new one.

The idea that PB systems costs are somehow normalized is erroneous, even when that would be the goal, I've yet to see a system that manges that.

[Segev]

Got it, that makes more sense. But I think bold is an assumption on your part that's never stated anywhere. The only way for class level A = class level B is if they're all the same, and that is the antithesis of depth. Even if that goal were somehow desirable, the amount of design time spent iterating to get there wouldn't be.

It is a tricky aspect of level and class design in a D&D-esq system, to be sure.

I actually think - despite the fact that I am sure this will be an unpopular opinion - that pre-3e had the right idea here: characters who have the same number of XP should be (roughly) equivalent in power, even if their actual class / character levels are wildly divergent. A level 7 rogue might be no more powerful than a level 4 wizard, but the rogue is level 7 because it took a LOT fewer XP to level up as a rogue than it does as a wizard. You could even take it to a custom-tailored extreme of including in each class table the XP value required to be at each level. i.e. the XP table is as much a part of every class table as the spell slot table is for spellcasting classes. The progression wouldn't even have to be smooth; a couple of "dead" levels that only give ribbons and some numbers to let you keep up the necessary hp or the like could be fast to power through, while "big money" levels with key class features might be gated behind larger gulfs of XP. Or, if you do make it smooth, you can use the dead levels as "filler" on the way to the big features.

This is, of course, potentially enormously complicated. It is simpler to have everyone at the same level; in fact, most games I'm in use milestone leveling at this point, because XP has become so pointless in D&D that it doesn't really make sense to bother with it. It has always been a bit of a pain to track, and even with the best of intentions and the most honest of players and DMs, discrepancies between what various players and the DM think the XP totals are arise. It's just hte nature of multiple people having to track things. (And if the DM is the only one tracking it, that doesn't eliminate the possibility that he's the one making mistakes.) However, these problems aren't necessarily too mcuh to deal with if XP actually contributes something...like being the REAL measure of power level.

This applies especially to 3.5, where XP was something you could spend on things other than leveling up. As long as what you spent it on kept you at the right power level for the number of XP you have, it wouldn't be a problem to spend it. And if a level 10 fighter is what it took to keep up with a level 5 wizard, then a level 10 fighter should have the same number of XP as a level 5 wizard.

When everyone's going to be at the same level, it is necessary to have every level be worth roughly the same amount across all classes. There can be a certain amount of averaging out, but it has to be pretty tightly done to prevent huge gaps in power between classes at various levels.

[PhoenixPhyre]

Got it, that makes more sense. But I think bold is an assumption on your part that's never stated anywhere. The only way for class level A = class level B is if they're all the same, and that is the antithesis of depth. Even if that goal were somehow desirable, the amount of design time spent iterating to get there wouldn't be.

My point is that they're not ever that way (equal in value). But multi classing treats them as if they are. Whether you take wizard 20 or barbarian 1, it costs one level. And that's absurd and makes it *bad* point buy. Because good point buy realizes that not everything is the same value and has different costs, often scaling ones. And good class/level doesn't care because barbarian 1 and wizard 20 are incomparable -- you can't get from one to the other. It's only level by level MC that allows that and causes the issue.


His point is that given there is zero granularity in the "point" you use to buy, which is the level up, all classes at all levels must be equally balanced for the system to be balanced.

That's wrong though, because it doesn't happen in any PB system I've played (about 5 or 6 that come to mind rite nao). For instance in WoD, if I've a character built for combat and have low mental stats, taking my first dot in academics probably won't be as useful as for someone who is playing a very intelligent hacker and takes their first dot in the same skill. So same cost for both, different value for cost across different characters. In the same vein, if the combat oriented character took their first dot in stealth, they are likely to get more out of it than from academics, for the simple reason that stealth is usually rolled with Dex, and as a combatant will likely have a decent Dex, different value for the cost across different skills.

And that's the "old paradigm" in the early 10s edition cost are not increasingly higher, so taking your first dot in a skill or your 5th cost the same, the value you get from your 5th point in anything is almost assuredly gonna be much higher than taking the first dot in anything, for the simple reason that you are getting better at one of your main schticks instead of getting a new one.

The idea that PB systems costs are somehow normalized is erroneous, even when that would be the goal, I've yet to see a system that manges that.

That's not my point. In fact, my point is rather the reverse. In a class-level system without level-by-level multiclassing, wizard X can be very different from barbarian X as long as the entire package from 1-X is relatively similar in overall strength[0]. Because by the time you get there, the system can assume you already have all the other levels before that. Things can progress at fairly different rates (some very smoothly like the rogue, others in larger jumps) and yet, on average, come out within the same ballpark. And since you can't directly choose between Wizard Y and Barbarian X, the direct level-by-level comparison is irrelevant. So you can't really compare wizard X to barbarian X directly. They're incomparable because they're not fungible. Level by level multiclassing breaks this because it lets you "jump trains" at no (direct) cost. You can take 1 level of fighter for the goodies, and then jump ship to wizard, paying only a small delay in your wizard features while still keeping all your fighter features. Which means you can compare Wizard X and Barbarian Y. And the game then presumes that those have the same cost (and thus value)...which is wrong. And will always be wrong. And must be wrong unless the classes are identical. And produces weirdness like Fighter 1/Wizard X =/= Wizard Y/Fighter 1/Wizard X-Y. To vet a character you need to know not only what levels they have, but what order they took them in. Sure, it's better than 3e (where you also needed to know what options and feats they took at each level and what their INT and CON were at each level so you can do skill points and decide whether they have the right number and haven't hit caps and had prereqs). But still it's not transitive. Which is stupid.

In a point-buy system[1], you do absolutely have "lock in", but you also (generally) have scaling or at least different costs[2]. So while picking up the first dot doesn't help you do "your thing" better, but is generally cheaper. And picking up <weaker thing> costs less than picking up <stronger thing>. Imagine if, in D&D context, learning a spell used a universal (per class) currency. And picking up a 9th level spell cost the same amount as 9 first level spells. Or maybe 18 first level spells. Currently, spell choices are "everything costs the same or is impossible to pick". Picking up a 9th level spell, as long as you can pick one, costs the same as picking a 1st level spell. But is exponentially more powerful. Sure, you've only got one slot, but you only have 4 1st level slots. And a 9th isn't 4x as strong as a 1st, it's tons and tons and tons stronger (almost a difference in kind as to what it can do). Which leads to the quadratic scaling[3] issue.

[0] or not--you could intentionally design for imbalance (such as that one ensemble game where mages >> everyone, but only one person plays a mage in any given session and mages mostly sit in their labs). But the key is that you're doing it intentionally and can design around that so it's fun regardless. That's not possible when you have escape routes to work around the design--then people just cherry pick the good and leave the balancing factors behind.
[1] one of the big complaints I've heard about WoD and the storyteller system in general is in fact that the point costs are all wonky. Which is doing class-based...just badly. Both sides can play at the "doing the other system badly" game, but neither should play, IMO.
[2] WoD at least does different costs at character creation than play, but I know things like M&M, HERO system, and GURPS actually have significantly different costs for various abilities and attributes. Adding <wide area> to something (making it a big aoe) may cost way more than adding <reskinnable> (making it look like whatever you want each time you use it). Or whatever. That lets you price things (more) accurately. Of course, point-buy systems also generally require very heavy, active GM involvement in pruning the possibility trees and whitelisting things/combinations (and banning other things).
[3] Personally, I think that the best scaling should be between linear and logarithmic. Rapid growth (both horizontal and vertical) early on, slowing down (but still positive slope) later on. So the gap between a 1st and 2nd level character is (hypothetically) 10 units, while the gap between a 19th and 20th level character is (hypothetically) 1 unit. Characters would "mature" early on (by level 5-7 ish) and then mainly "fill out" rather than vaulting higher and higher to the point the system falls apart. Because systems don't handle power-law growth very well. Bounded accuracy or not, the actual system bounds are fairly narrow and rigid. The actual absolute power levels aren't--you can make a game that handles god-caliber creatures just fine. As long as you don't expect it to handle normal folks as more than just scene dressing. And vice versa. But handling the whole scope from normal joe to god-eater in a single game doesn't work unless you abstract things so far that the only difference is in how you describe stuff. Which, I guess, is another option. Just one I don't like.

[Psyren]

My point is that they're not ever that way (equal in value). But multi classing treats them as if they are. Whether you take wizard 20 or barbarian 1, it costs one level. And that's absurd and makes it *bad* point buy. Because good point buy realizes that not everything is the same value and has different costs, often scaling ones. And good class/level doesn't care because barbarian 1 and wizard 20 are incomparable -- you can't get from one to the other. It's only level by level MC that allows that and causes the issue.

This is our impasse - you think the benefits gained from allowing this kind of multiclassing are not worth any issues that might arise, I do. And WotC clearly does too, otherwise MC (already an optional rule) would not be allowed in sanctioned play. So we may have to simply agree to disagree.


It is a tricky aspect of level and class design in a D&D-esq system, to be sure.

I actually think - despite the fact that I am sure this will be an unpopular opinion - that pre-3e had the right idea here: characters who have the same number of XP should be (roughly) equivalent in power, even if their actual class / character levels are wildly divergent. A level 7 rogue might be no more powerful than a level 4 wizard, but the rogue is level 7 because it took a LOT fewer XP to level up as a rogue than it does as a wizard. You could even take it to a custom-tailored extreme of including in each class table the XP value required to be at each level. i.e. the XP table is as much a part of every class table as the spell slot table is for spellcasting classes. The progression wouldn't even have to be smooth; a couple of "dead" levels that only give ribbons and some numbers to let you keep up the necessary hp or the like could be fast to power through, while "big money" levels with key class features might be gated behind larger gulfs of XP. Or, if you do make it smooth, you can use the dead levels as "filler" on the way to the big features.

This is, of course, potentially enormously complicated. It is simpler to have everyone at the same level; in fact, most games I'm in use milestone leveling at this point, because XP has become so pointless in D&D that it doesn't really make sense to bother with it. It has always been a bit of a pain to track, and even with the best of intentions and the most honest of players and DMs, discrepancies between what various players and the DM think the XP totals are arise. It's just hte nature of multiple people having to track things. (And if the DM is the only one tracking it, that doesn't eliminate the possibility that he's the one making mistakes.) However, these problems aren't necessarily too mcuh to deal with if XP actually contributes something...like being the REAL measure of power level.

This applies especially to 3.5, where XP was something you could spend on things other than leveling up. As long as what you spent it on kept you at the right power level for the number of XP you have, it wouldn't be a problem to spend it. And if a level 10 fighter is what it took to keep up with a level 5 wizard, then a level 10 fighter should have the same number of XP as a level 5 wizard.

When everyone's going to be at the same level, it is necessary to have every level be worth roughly the same amount across all classes. There can be a certain amount of averaging out, but it has to be pretty tightly done to prevent huge gaps in power between classes at various levels.

The gaps are really not that huge. Or rather, they would be if this was a PvP game and needed to be balanced for duels and the like, but it's not.

[Rukelnikov]

My point is that they're not ever that way (equal in value). But multi classing treats them as if they are. Whether you take wizard 20 or barbarian 1, it costs one level. And that's absurd and makes it *bad* point buy. Because good point buy realizes that not everything is the same value and has different costs, often scaling ones. And good class/level doesn't care because barbarian 1 and wizard 20 are incomparable -- you can't get from one to the other. It's only level by level MC that allows that and causes the issue.



That's not my point. In fact, my point is rather the reverse. In a class-level system without level-by-level multiclassing, wizard X can be very different from barbarian X as long as the entire package from 1-X is relatively similar in overall strength[0]. Because by the time you get there, the system can assume you already have all the other levels before that. Things can progress at fairly different rates (some very smoothly like the rogue, others in larger jumps) and yet, on average, come out within the same ballpark. And since you can't directly choose between Wizard Y and Barbarian X, the direct level-by-level comparison is irrelevant. So you can't really compare wizard X to barbarian X directly. They're incomparable because they're not fungible. Level by level multiclassing breaks this because it lets you "jump trains" at no (direct) cost. You can take 1 level of fighter for the goodies, and then jump ship to wizard, paying only a small delay in your wizard features while still keeping all your fighter features. Which means you can compare Wizard X and Barbarian Y. And the game then presumes that those have the same cost (and thus value)...which is wrong. And will always be wrong. And must be wrong unless the classes are identical. And produces weirdness like Fighter 1/Wizard X =/= Wizard Y/Fighter 1/Wizard X-Y. To vet a character you need to know not only what levels they have, but what order they took them in. Sure, it's better than 3e (where you also needed to know what options and feats they took at each level and what their INT and CON were at each level so you can do skill points and decide whether they have the right number and haven't hit caps and had prereqs). But still it's not transitive. Which is stupid.

In a point-buy system[1], you do absolutely have "lock in", but you also (generally) have scaling or at least different costs[2]. So while picking up the first dot doesn't help you do "your thing" better, but is generally cheaper. And picking up <weaker thing> costs less than picking up <stronger thing>. Imagine if, in D&D context, learning a spell used a universal (per class) currency. And picking up a 9th level spell cost the same amount as 9 first level spells. Or maybe 18 first level spells. Currently, spell choices are "everything costs the same or is impossible to pick". Picking up a 9th level spell, as long as you can pick one, costs the same as picking a 1st level spell. But is exponentially more powerful. Sure, you've only got one slot, but you only have 4 1st level slots. And a 9th isn't 4x as strong as a 1st, it's tons and tons and tons stronger (almost a difference in kind as to what it can do). Which leads to the quadratic scaling[3] issue.

[0] or not--you could intentionally design for imbalance (such as that one ensemble game where mages >> everyone, but only one person plays a mage in any given session and mages mostly sit in their labs). But the key is that you're doing it intentionally and can design around that so it's fun regardless. That's not possible when you have escape routes to work around the design--then people just cherry pick the good and leave the balancing factors behind.
[1] one of the big complaints I've heard about WoD and the storyteller system in general is in fact that the point costs are all wonky. Which is doing class-based...just badly. Both sides can play at the "doing the other system badly" game, but neither should play, IMO.
[2] WoD at least does different costs at character creation than play, but I know things like M&M, HERO system, and GURPS actually have significantly different costs for various abilities and attributes. Adding <wide area> to something (making it a big aoe) may cost way more than adding <reskinnable> (making it look like whatever you want each time you use it). Or whatever. That lets you price things (more) accurately. Of course, point-buy systems also generally require very heavy, active GM involvement in pruning the possibility trees and whitelisting things/combinations (and banning other things).
[3] Personally, I think that the best scaling should be between linear and logarithmic. Rapid growth (both horizontal and vertical) early on, slowing down (but still positive slope) later on. So the gap between a 1st and 2nd level character is (hypothetically) 10 units, while the gap between a 19th and 20th level character is (hypothetically) 1 unit. Characters would "mature" early on (by level 5-7 ish) and then mainly "fill out" rather than vaulting higher and higher to the point the system falls apart. Because systems don't handle power-law growth very well. Bounded accuracy or not, the actual system bounds are fairly narrow and rigid. The actual absolute power levels aren't--you can make a game that handles god-caliber creatures just fine. As long as you don't expect it to handle normal folks as more than just scene dressing. And vice versa. But handling the whole scope from normal joe to god-eater in a single game doesn't work unless you abstract things so far that the only difference is in how you describe stuff. Which, I guess, is another option. Just one I don't like.

That's the same thing I said.

And if the GM has to actively prune "combos" in point buy systems, why can't they do it in DnD? Which has orders of magnitudes less combinations than either WoD or M&M (haven't lpayed HERO). M&M 2e btw is broken beyond belief in the disparity of power per cost that can be bought, and you've already said you don't think WoD does it correctly, so my question becomes, which system does? Because that was also the conclusion of my argument, NO system does it "properly" but most do it well enough, 5e included.

[Segev]

The gaps are really not that huge. Or rather, they would be if this was a PvP game and needed to be balanced for duels and the like, but it's not.

A valid argument. I don't care enough to dispute the assertion, and it is, if true, a valid reason to discard XP entirely and just go with milestone leveling.

This does mean that, if I understand you correctly, you're taking the position that all of the classes are more or less even at any particular level?

[Psyren]

A valid argument. I don't care enough to dispute the assertion, and it is, if true, a valid reason to discard XP entirely and just go with milestone leveling.

This does mean that, if I understand you correctly, you're taking the position that all of the classes are more or less even at any particular level?

I am taking the position that a party of any 4 classes with tier-appropriate gear can succeed against a challenge of the appropriate CR. That is the explicit standard for this game, and it doesn't require every individual class level to equal every other individual class level.

[PhoenixPhyre]

I am taking the position that a party of any 4 classes with tier-appropriate gear can succeed against a challenge of the appropriate CR. That is the explicit standard for this game, and it doesn't require every individual class level to equal every other individual class level.

And you know what? I basically agree with this. And "tier appropriate gear" isn't all that much (a common magic sword does the trick for most of it. That's the game's real mechanical baseline.

The issue for me is that in games where multi classing is normal and optimization is expected, you can't do that. And you have to throw out any idea of CR appropriate encounters if you don't want endless curb stomps without any challenge. That is, the extra synergy and power from multi classing makes it an arms race. If anyone optimizes, everyone must, including the DM, or things go poorly. It raises the floor substantially. And in ways that make most of the other systems of the game either redundant or oddly performing. And makes reasonable worlds that aren't totally gonzo (dripping in world-ending threats with constant churn of demon princes and the like) impossible. Basically, I see the high end as being higher than the system or fictional base can really handle well, while the regular part works pretty fine.

[sithlordnergal]

And you know what? I basically agree with this. And "tier appropriate gear" isn't all that much (a common magic sword does the trick for most of it. That's the game's real mechanical baseline.

The issue for me is that in games where multi classing is normal and optimization is expected, you can't do that. And you have to throw out any idea of CR appropriate encounters if you don't want endless curb stomps without any challenge. That is, the extra synergy and power from multi classing makes it an arms race. If anyone optimizes, everyone must, including the DM, or things go poorly. It raises the floor substantially. And in ways that make most of the other systems of the game either redundant or oddly performing. And makes reasonable worlds that aren't totally gonzo (dripping in world-ending threats with constant churn of demon princes and the like) impossible. Basically, I see the high end as being higher than the system or fictional base can really handle well, while the regular part works pretty fine.

See...I don't actually think that's as much as an issue as you think. The nice part about 5e is its pretty easy to scale encounters up and down. The only time it becomes difficult is if there's only one or two highly optimized characters in a low-optimization party, or vice versa. I don't think it becomes an "arms race" by any means. And heck, you don't always have to scale an encounter up or down, just go in knowing that the party will likely be able to deal with this fairly easily. Nothing wrong with a powerful party occasionally dealing with an easy threat.

Having higher scaling encounters also doesn't make reasonable worlds impossible. You don't need to throw a constant churn of liches, demon lords, and other world ending threats to give such parties a challenging encounter. Just use monsters in different ways. Wanna challenge a party with a Necromancer? Give him a Ghost or Wraith, a Poltergeist, and a single Ogre Zombie. Give the Ogre Zombie Plate Armor and a Shield, set it in the middle of a hallway that the party has to go down, have it only Dodge, put the Necromancer behind it, and boom. You now have a dangerous encounter.

The Wraith/Ghost can use Hit-and-Run tactics by slipping past walls, forcing players to use Readied Actions or Attacks of Opportunity to kill them, the Ogre Zombie is now a giant roadblock that the party has to find a way past, the Necromancer can bombard the party from behind cover and step around a corner to avoid being targeted by things, and the Poltergeist is there to cause problems by shoving people around. If you really want to be evil, replace the Poltergeist with a Bodak hidden behind a wall with just enough holes in it that it doesn't count as total cover, that way the party has to deal with their aura.

[PhoenixPhyre]

See...I don't actually think that's as much as an issue as you think. The nice part about 5e is its pretty easy to scale encounters up and down. The only time it becomes difficult is if there's only one or two highly optimized characters in a low-optimization party, or vice versa. I don't think it becomes an "arms race" by any means. And heck, you don't always have to scale an encounter up or down, just go in knowing that the party will likely be able to deal with this fairly easily. Nothing wrong with a powerful party occasionally dealing with an easy threat.

Having higher scaling encounters also doesn't make reasonable worlds impossible. You don't need to throw a constant churn of liches, demon lords, and other world ending threats to give such parties a challenging encounter. Just use monsters in different ways. Wanna challenge a party with a Necromancer? Give him a Ghost or Wraith, a Poltergeist, and a single Ogre Zombie. Give the Ogre Zombie Plate Armor and a Shield, set it in the middle of a hallway that the party has to go down, have it only Dodge, put the Necromancer behind it, and boom. You now have a dangerous encounter.

The Wraith/Ghost can use Hit-and-Run tactics by slipping past walls, forcing players to use Readied Actions or Attacks of Opportunity to kill them, the Ogre Zombie is now a giant roadblock that the party has to find a way past, the Necromancer can bombard the party from behind cover and step around a corner to avoid being targeted by things, and the Poltergeist is there to cause problems by shoving people around. If you really want to be evil, replace the Poltergeist with a Bodak hidden behind a wall with just enough holes in it that it doesn't count as total cover, that way the party has to deal with their aura.

What I meant by "arms race" is that if anyone in the party optimizes, either everyone does or there's issues. And the DM has to treat it much differently--more like a chess game divorced from anything in the setting and choosing monsters and tactics purely on the mechanical aspects, not "what makes sense in the setting".

[sithlordnergal]

What I meant by "arms race" is that if anyone in the party optimizes, either everyone does or there's issues. And the DM has to treat it much differently--more like a chess game divorced from anything in the setting and choosing monsters and tactics purely on the mechanical aspects, not "what makes sense in the setting".

Fair, if only one person is highly optimized it can make creating encounters annoying. But then, that's not really a problem with multiclassing, that's just an issue with optimization in general. And optimization is always going to exist, no matter what you do. Heck, I can name two times where a character was optimized to the point that it caused issues due to them being too strong. Both times they were single classed characters, and both times we had multiclass characters in the party. One was a Moon Druid in Tier 1, which is a well known issue and was solved once the party reached Tier 2. The other was an Artificer, which is the only character I have ever needed to ask a player to change because they were so much stronger than the rest of the party.

I mean...yes and no? It does become more like a chess game, I won't deny that, but I also find that a ton of fun. My favorite part of DMing is that chess-like encounter building where you make an encounter that drives the players to the brink of their strategic and creative ability, and forces them to come at a problem in different ways, cause trying to brute force it will only result in death. And you can do that and still have it make sense in the setting.

Goblins and Kobolds aren't mindless idiots that just charge at a foe, they're sneaky and devious. They're not gonna fight face to face, they're gonna use traps, and kill zones where they can shoot you from full cover. Orcs may have a low intelligence, but they're typically portrayed with a rough, war-like culture. You better believe they're experts when it comes to war tactics. Undead run the gambit of mindless to brilliant, and the mindless ones are typically controlled by one of those genius ones. The Ogre Zombie isn't the one that had the idea of wearing armor and dodging, its just following the only command it Master gave it "Stand here and Dodge". Wolves are shown to use basic ambush, flanking, and pack tactics IRL, why can't they do so in DnD?

Now, obviously if you dislike making and running those chess-like encounters, then the above stuff isn't going to be fun to do. And if you're not using those chess-like encounters, then I'm willing to bet the disparity between optimized and unoptimized characters is going to be a lot more pronounced. Though as I said before, issues caused by optimization don't typically come from multiclassed characters in my experience. They come from OP highly optimized single classed characters.


EDIT: Correction, I can name 3 times when an optimized character caused problems. The first two were already listed, the last one was when I was playing a Moon Druid in Tomb of Annihilation. However, that was partially due to me playing a Moon Druid, and partially because the DM allowed me to choose what animals I summoned whenever I cast Conjure Animals...and 8 Velociraptors are a serious threat when they gang up on a single creature with no AoEs.

[PhoenixPhyre]

Fair, if only one person is highly optimized it can make creating encounters annoying. But then, that's not really a problem with multiclassing, that's just an issue with optimization in general. And optimization is always going to exist, no matter what you do. Heck, I can name two times where a character was optimized to the point that it caused issues due to them being too strong. Both times they were single classed characters, and both times we had multiclass characters in the party. One was a Moon Druid in Tier 1, which is a well known issue and was solved once the party reached Tier 2. The other was an Artificer, which is the only character I have ever needed to ask a player to change because they were so much stronger than the rest of the party.

I mean...yes and no? It does become more like a chess game, I won't deny that, but I also find that a ton of fun. My favorite part of DMing is that chess-like encounter building where you make an encounter that drives the players to the brink of their strategic and creative ability, and forces them to come at a problem in different ways, cause trying to brute force it will only result in death. And you can do that and still have it make sense in the setting.

Goblins and Kobolds aren't mindless idiots that just charge at a foe, they're sneaky and devious. They're not gonna fight face to face, they're gonna use traps, and kill zones where they can shoot you from full cover. Orcs may have a low intelligence, but they're typically portrayed with a rough, war-like culture. You better believe they're experts when it comes to war tactics. Undead run the gambit of mindless to brilliant, and the mindless ones are typically controlled by one of those genius ones. The Ogre Zombie isn't the one that had the idea of wearing armor and dodging, its just following the only command it Master gave it "Stand here and Dodge". Wolves are shown to use basic ambush, flanking, and pack tactics IRL, why can't they do so in DnD?

Now, obviously if you dislike making and running those chess-like encounters, then the above stuff isn't going to be fun to do. And if you're not using those chess-like encounters, then I'm willing to bet the disparity between optimized and unoptimized characters is going to be a lot more pronounced. Though as I said before, issues caused by optimization don't typically come from multiclassed characters in my experience. They come from OP highly optimized single classed characters.


EDIT: Correction, I can name 3 times when an optimized character caused problems. The first two were already listed, the last one was when I was playing a Moon Druid in Tomb of Annihilation. However, that was partially due to me playing a Moon Druid, and partially because the DM allowed me to choose what animals I summoned whenever I cast Conjure Animals...and 8 Velociraptors are a serious threat when they gang up on a single creature with no AoEs.

If I want chess-like encounters, I'd play a tactics video game. They can do it way better and don't have to worry about things like mechanical complexity or other such things. Doing that in a TTRP seems like a total waste of everyone's time, since neither the game engine nor the players can take advantage of it.

Multiclassing takes optimization and puts it in overdrive specifically because it add a bunch of unintended options. Single-classed optimization tends to focus on well-known "broken" things (ie summoning). And Moon Druid isn't really a problem even in T1...in part because T1 lasts for what, 5 sessions max?

What I object to is always having to think in that mode as soon as one person decides to optimize. Suddenly the vast majority of common tactics and "fluff-appropriate" compositions go out the window--they're curbstomps by default. Instead, you have to "play everything smart". Which gets in the way of a lot of things and reduces the game to mere "find appropriate button, press appropriate button". And strongly favors spell-casters. In fact, it turbocharges the value of the versatility and "it just happens" that spellcasters bring to the table. None of which make much sense or fun for the rest of the party.

And when the existence of a single optimized player character warps the entire game around it so now every encounter that is expected to be any kind of a challenge has to intentionally neutralize that one player, the system is broken. You've passed what the system was intended to do. And now instead of a team game, it's "watch so-and-so shine while we all cower here." Or if everyone's optimized, then you end up having to play completely broken and degenerate (meaning there's a "right way" and then there's all the ways that don't work) solved game.

Just like sure, you can play tennis without a net and with randomly exploding tennis balls. And some people might have fun. But it's hardly a game of tennis any more. That's how optimized D&D feels.

[Thunderous Mojo]

And when the existence of a single optimized player character warps the entire game around it so now every encounter that is expected to be any kind of a challenge has to intentionally neutralize that one player, the system is broken. You've passed what the system was intended to do. And now instead of a team game, it's "watch so-and-so shine while we all cower here." Or if everyone's optimized, then you end up having to play completely broken and degenerate (meaning there's a "right way" and then there's all the ways that don't work) solved game.

Just like sure, you can play tennis without a net and with randomly exploding tennis balls. And some people might have fun. But it's hardly a game of tennis any more. That's how optimized D&D feels.

Part of the ebb and flow of a longterm D&D game, in my experience, is sometimes a situation allows a fellow PC to shine, and puts your own PC at a Disadvantage, and sometimes the inverse is true.

The Icewind Dale module has an Encounter, that includes a CR 6 Frost Giant Skeleton, that can be facedby a second character level adventuring group. The Giant Skeleton does an average of 25 points of damage with their axe, or has a chilling stare that does 35 points of damage on a failed DC 15 Constitution save.

A D&D Party with a Cleric, is going to likely have a much different experience than a party of Rogues, that is trying to snipe and kite the giant skeleton because they can not survive an axe blow, let alone the Magical Gaze attack from the undead giant.

Would you consider the cleric, “optimized” in this type of scenario?

“Optimized” is an empty term, and I wish people would define what they intend to convey, when they use it. One ‘Optimizes’ to reach a desired result. A cleric optimized to be the ‘Face’ of the party, is going to play differently compared to a cleric designed to maximize a Spirit Guardians spell.

[Witty Username]

The issue for me is that in games where multi classing is normal and optimization is expected, you can't do that. And you have to throw out any idea of CR appropriate encounters if you don't want endless curb stomps without any challenge.

I don't think that is an issue with multiclassing so much as a system expectation in 5e. From what I have been lead to believe from the DMG and my personal experience, any encounter below deadly should be considered a guaranteed victory by the party. With encounters medium and easy encounters being finishable without resources not considered unusual (easy calls out that losing HP, should be considered a possibility, in line with a worst case scenario). So if we want challenging encounters we already need to toss the idea of CR appropriate encounters.

[tiornys]

I don't think that is an issue with multiclassing so much as a system expectation in 5e. From what I have been lead to believe from the DMG and my personal experience, any encounter below deadly should be considered a guaranteed victory by the party. With encounters medium and easy encounters being finishable without resources not considered unusual (easy calls out that losing HP, should be considered a possibility, in line with a worst case scenario). So if we want challenging encounters we already need to toss the idea of CR appropriate encounters.
Encounters below deadly should indeed be (nearly) guaranteed victory, but there's a huge difference between victory with expected cost of 20% of the party's daily resources and victory with expected cost of 2% of the party's daily resources. The former party will find themselves challenged and stretched by a dungeon crawl consisting of 5-7 encounters; the latter will not.

For me, the important metric by which to measure the impact of character to character imbalance on a cooperative RPG is this: how easily can two players, lacking system mastery, build characters with reasonable appearing choices and end up with widely disparate power levels? And especially, how easily can one character build unintentionally end up outshining another character build in nearly all aspects of the game? It is not as easy to do this in 5E as it was in 3E, but it's still much too easy for me to consider 5E to be reasonably balanced.

[Witty Username]

Encounters below deadly should indeed be (nearly) guaranteed victory, but there's a huge difference between victory with expected cost of 20% of the party's daily resources and victory with expected cost of 2% of the party's daily resources. The former party will find themselves challenged and stretched by a dungeon crawl consisting of 5-7 encounters; the latter will not.

See that's the thing though, a medium encounter is already the 2% of the parties resources at that point. And Hard is not much more than that from personal experience, the 20% is when we are talking about deadly encounters which is already past the system expectations for a 6 to 8 encounter day like the DMG recommends for medium and hard encounters.

[tiornys]

See that's the thing though, a medium encounter is already the 2% of the parties resources at that point. And Hard is not much more than that from personal experience, the 20% is when we are talking about deadly encounters which is already past the system expectations for a 6 to 8 encounter day like the DMG recommends for medium and hard encounters.
System expectation is for medium and hard encounters to exhaust between 10%-20% of the party's resources. My limited experience with casually built characters suggests this is reasonably accurate. I don't think it takes that much optimization to shift that expectation up a category (hard to deadly encounters should exhaust between 10%-20% of the party's resources), and that sounds roughly like the experience you've mentioned. Mid-high to high op characters can routinely take on (baseline) deadly encounters with something like 2% resources spent--it takes deadly++ or higher levels of challenge to start significantly taxing their resources, thus the line about throwing out CR entirely.

[LibraryOgre]

It does address the complaints made about level-by-level multiclassing. You pick what you're going to do at chargen, and you pay for the increased versatility with much less power. You're not dipping anything for a shallow investment, either; you commit to the multiclass or you don't do it.

Which is why I tend to like subclasses for multiclassing (with feats for garnish); effectively, it can mimic that same effect by providing a partial class to add to your main class.

In some ways, it winds up similar to 1e multiclassing, with its lower level limits. If I was a half-orc cleric/assassin, my cleric level limitations meant that, for a good portion of my career, I was an assassin with a bit of cleric ability (including the fact that 1e clerics could use any weapon, if multiclassed). I'd never be the cleric that a full cleric was, but I would be far more of a cleric than any other assassin.

[Psyren]

Fair, if only one person is highly optimized it can make creating encounters annoying. But then, that's not really a problem with multiclassing, that's just an issue with optimization in general.

Totally agreed. People on different optimization levels in the party will cause issues whether multiclassing is possible or not.


I mean...yes and no? It does become more like a chess game, I won't deny that, but I also find that a ton of fun. My favorite part of DMing is that chess-like encounter building where you make an encounter that drives the players to the brink of their strategic and creative ability, and forces them to come at a problem in different ways, cause trying to brute force it will only result in death. And you can do that and still have it make sense in the setting.

Indeed. Not every encounter needs to be a finely tuned puzzle, but some should be. (And I personally love chess puzzles, especially the ones with multiple solutions.)

[sithlordnergal]

If I want chess-like encounters, I'd play a tactics video game. They can do it way better and don't have to worry about things like mechanical complexity or other such things. Doing that in a TTRP seems like a total waste of everyone's time, since neither the game engine nor the players can take advantage of it.

I mean, I've found 5e does a really good job with chess-like encounters. Its just a matter of how much work the DM is willing to put into the encounter creation and balancing. Even something small can make a difference. There's a massive difference between fighting Kobolds in a cave, and fighting Kobolds in a cave that have set traps around every corner, and are literally hiding in the walls to shoot you from the safety of full cover. The first one is just an encounter where you need to kill the Kobolds, the second one requires strategy. Though again, that is fully on DM and player preferences. I suspect you'd deeply dislike playing in my games, because a majority of encounters require that sort of strategizing to survive.



Multiclassing takes optimization and puts it in overdrive specifically because it add a bunch of unintended options. Single-classed optimization tends to focus on well-known "broken" things (ie summoning). And Moon Druid isn't really a problem even in T1...in part because T1 lasts for what, 5 sessions max?

I mean, yes multiclassing does give more options to optimize, but those optimizations are not that much stronger than single classed characters. They have a bit more versatility, yes, but not so much versatility or power that they go far above the capabilities of a single class. I tend to find single classed optimization to be far more game breaking because they hyper focus on those game breaking options. And unless you completely ban certain things, such as Animate Dead and summoning spells, or about half of an Artificer's Infusion list, they're far harder to bring into line. Which, sure, you can do that...but I've never really been a fan of nerfs and bans. I personally feel if I have to ban something, then I've done a poor job as a DM by failing to learn the system I am DMing for.

Heck, the only character I've had to make a player swap out due to being OP compared to the rest of the party was a single classed Artificer. This was in a party of multiclassed characters, with a range of optimization. And even then, I'm not going to ban the Artificer class. I'm going to learn the in's and out's of Artificer so I can be better prepared to run for them. Once I've identified what they're amazing at and what they're bad at, I'll be good to go. I'll be able to design encounters based around their strengths and weaknesses, just like I do for every single other class in the game.




What I object to is always having to think in that mode as soon as one person decides to optimize. Suddenly the vast majority of common tactics and "fluff-appropriate" compositions go out the window--they're curbstomps by default. Instead, you have to "play everything smart". Which gets in the way of a lot of things and reduces the game to mere "find appropriate button, press appropriate button". And strongly favors spell-casters. In fact, it turbocharges the value of the versatility and "it just happens" that spellcasters bring to the table. None of which make much sense or fun for the rest of the party.


I mean, that's just a general dislike of optimization. Which is fine, you're free to dislike optimization, but I have a few issues with your stance:

1) You're basically stating that we should remove multiclassing because that will solve the issue of optimization, this is simply not true. I typically find multiclassed characters to be in the same general power level of single classed characters, and when you get into hyper precise optimization, single classed characters tend to be better than multiclassed ones.


2) Optimization will always exist. Even if you took away Feats, Mutliclassing, every single Spell, made every Race exactly the same, it will always exist. That's just the nature of having things that can be compared to each other. The only way you'd avoid having that is if you make every class essentially the same...which from what I hear, is something 4e kind of did. And lets face it...4e was a biiiit unpopular when compared to 3.x and 5e.


3) Is that the Stormwind Fallacy I see?


Suddenly the vast majority of common tactics and "fluff-appropriate" compositions go out the window--they're curbstomps by default. Instead, you have to "play everything smart". Which gets in the way of a lot of things and reduces the game to mere "find appropriate button, press appropriate button".

That is the Stormwind Fallacy! Optimization doesn't force fluff-appropriate combinations out the window, it doesn't force you to play "everything smart", nor does it get in the way of things or reduce it to a mere "find appropriate button to press". You can 100% have all of that even as you optimize. Heck, multiclassing can help encourage fluff-appropriate choices, by making those choices even more applicable or useful. Obviously you can also get the "multiclass just for optimization", but you also have the "multiclass because fluff".


4) Versatility is always going to be valued. Heck, versatility is going to be even more valued in a game where versatility is artificially limited. A wizard taking the Fly spell has suddenly added a ton of versatility to their spell list, but that in and of itself is not optimization. And Martials have plenty of versatility of their own. Sure, they can't cast spells like fly, but they typically excel at skill based checks. If you're talking to someone, you're going to want the person who's really good at persuasion, deception, or intimidation checks. A wizard or bard can try to use Charm Person, but its a bad idea since everyone knows you cast Charm Person, it doesn't actually do anything other than give Advantage on the check, and once the Charm wears off, the charmed person is generally going to be hostile to you cause they know you charmed them.




And when the existence of a single optimized player character warps the entire game around it so now every encounter that is expected to be any kind of a challenge has to intentionally neutralize that one player, the system is broken. You've passed what the system was intended to do. And now instead of a team game, it's "watch so-and-so shine while we all cower here." Or if everyone's optimized, then you end up having to play completely broken and degenerate (meaning there's a "right way" and then there's all the ways that don't work) solved game.

Just like sure, you can play tennis without a net and with randomly exploding tennis balls. And some people might have fun. But it's hardly a game of tennis any more. That's how optimized D&D feels.

Again, optimization is always going to exist. Multiclassing has very little to do with how strong an optimized character is, and no matter what you do it will always exist. Even if you were to remove multiclassing and feats, I would personally still want to make a highly optimized character to the best of my ability, because I find that fun. I have fun calculating the very best combination of things, and comparing the different subclasses to find the most optimized option to do some thing. Doesn't matter if that "thing" is damage, AC, HP, skill checks, or movement speed. And the same holds true for pretty much every other optimizer. The only way you'd get rid of optimization is if you removed optimizers from the game entirely, and I'm not going to be giving up DnD any time soon.

And finally, you're completely incorrect about what happens if you have an optimized party. Having an entire table of optimized characters doesn't "break the game", nor does it mean there's a "right way and all the ways that don't work". That is simply not true. Them being able to handle higher CR challenges does not mean you have to play in some "completely broken and degenerate way". You can still challenge them with lower CR foes. You just need to be more tactical with those foes. And if your idea of tactics is "there's only one way to solve this"...well, you need to learn how to make better challenges. A good challenge has multiple solutions, and the best ones have more options than just "use magic".

[Segev]

Which is why I tend to like subclasses for multiclassing (with feats for garnish); effectively, it can mimic that same effect by providing a partial class to add to your main class.

In some ways, it winds up similar to 1e multiclassing, with its lower level limits. If I was a half-orc cleric/assassin, my cleric level limitations meant that, for a good portion of my career, I was an assassin with a bit of cleric ability (including the fact that 1e clerics could use any weapon, if multiclassed). I'd never be the cleric that a full cleric was, but I would be far more of a cleric than any other assassin.

It does require either that you abandon any concept that isn't super-vanilla in both the base and the "multi-class subclass," or that you have subclasses that are half a subclass for the base class AND half a subclass for the "multi-class" they're subbing for, though.

If the only way to play a cleric/wizard is to be an Arcana Cleric or a Divine Wizard, you can't be a Cleric of War who studies War Magic. Nor can you be a Healer Cleric who also studies the magic of Necromancy to become a master of Life and Death.

It can be a valid build choice, but it doesn't do the whole job. Neither would having no subclasses-as-multiclass options and only having multiclassing do the whole job. They serve different kinds of concepts.

[sithlordnergal]

It does require either that you abandon any concept that isn't super-vanilla in both the base and the "multi-class subclass," or that you have subclasses that are half a subclass for the base class AND half a subclass for the "multi-class" they're subbing for, though.

If the only way to play a cleric/wizard is to be an Arcana Cleric or a Divine Wizard, you can't be a Cleric of War who studies War Magic. Nor can you be a Healer Cleric who also studies the magic of Necromancy to become a master of Life and Death.

It can be a valid build choice, but it doesn't do the whole job. Neither would having no subclasses-as-multiclass options and only having multiclassing do the whole job. They serve different kinds of concepts.

This right here. Single classes, subclasses, and feats simply can't cover every single potential option for a possible character. You would need an absolutely flood of subclasses to get close to it. That's why I say that just using Subclasses and Feats to emulate multiclassing would do a very, very poor job. You'd end up with a half-assed system where you don't actually get to make the characters you really want to make.

[stoutstien]

This right here. Single classes, subclasses, and feats simply can't cover every single potential option for a possible character. You would need an absolutely flood of subclasses to get close to it. That's why I say that just using Subclasses and Feats to emulate multiclassing would do a very, very poor job. You'd end up with a half-assed system where you don't actually get to make the characters you really want to make.

You don't need a system. You need a template. The player describes what they want to play and they work with them DM to create it.

[animorte]

I don't understand why any one class needs to do the full job of their entire class and successfully emulate entirely what another class is capable of. We already have a fair amount of overlap and, quite frankly, a player that comes into the game trying to do everybody else's job is a part of the problem.

I understand certain situations when you only have 1-3 players and all pillars of the game and roles of adventure are much more difficult to account for specifically because you don't have enough players. I also kind of get it when you can't really trust the other players around the table or you can't really trust the DM to create a balanced atmosphere to prepare for everybody to feel useful. Kind of, on that last bit, just because that sounds more like an issue with the people than it is with the game.

I appreciate people trying to conjure up a more versatile build path for everything, but I don't really understand the route that is being taken here to theory-craft it.

Edit: I have a couple examples: A couple months ago, in homebrew I posted the initial working concept (https://forums.giantitp.com/showthread.php?647451-Build-your-own-subclass!) of a potential system that would fully allow someone to pull a different subclass feature from another subclass entirely at the relevant level. (The majority of response preferred to treat it like multi-classing, starting features from earliest levels.)

I've also run a couple classless games that allow one to select features, skills, proficiencies, etc. upon leveling up and that's worked well enough to create this ongoing concept we well.

[sithlordnergal]

You don't need a system. You need a template. The player describes what they want to play and they work with them DM to create it.

I mean, you can do that...but multiclassing serves the same purpose and is far easier than having to make a homebrew class every time a player wants to play a Cleric/Wizard. Its far easier to just say "Play Cleric/Wizard" in order to achieve what the player wants to play over having to pick and choose what Cleric abilities the Wizard should get in order to replicate what the player wants to play.

[stoutstien]

I mean, you can do that...but multiclassing serves the same purpose and is far easier than having to make a homebrew class every time a player wants to play a Cleric/Wizard. Its far easier to just say "Play Cleric/Wizard" in order to achieve what the player wants to play over having to pick and choose what Cleric abilities the Wizard should get in order to replicate what the player wants to play.

Seeing how most of what players are seeking out of that combination is circumventing opportunity costs of being one or the other Id say multiclassing doesn't serve it's purpose for my games. Easier sure but based on the time players tend to spend with a PC that's a poor excuse to go that route. No reason to rush the foundation.

If a player came to me with a sound reason to mix the two then I might as well take the time to come up with something that works rather than use a janky patchwork of two classes being treated as piles of mechanics.

[PhoenixPhyre]

This right here. Single classes, subclasses, and feats simply can't cover every single potential option for a possible character. You would need an absolutely flood of subclasses to get close to it. That's why I say that just using Subclasses and Feats to emulate multiclassing would do a very, very poor job. You'd end up with a half-assed system where you don't actually get to make the characters you really want to make.

Here's the thing. The system never promises "every single potential option". It says "here are the supported archetypes. Pick one." That's the purpose of a class-level system--to restrict choices to the ones provided.

You want a point-buy system. I don't. D&D is not one. Nor would making it a half-baked point-buy system be to its benefit.


I don't understand why any one class needs to do the full job of their entire class and successfully emulate entirely what another class is capable of. We already have a fair amount of overlap and, quite frankly, a player that comes into the game trying to do everybody else's job is a part of the problem.

I understand certain situations when you only have 1-3 players and all pillars of the game and roles of adventure are much more difficult to account for specifically because you don't have enough players. I also kind of get it when you can't really trust the other players around the table or you can't really trust the DM to create a balanced atmosphere to prepare for everybody to feel useful. Kind of, on that last bit, just because that sounds more like an issue with the people than it is with the game.


Exactly. And that second paragraph? If you can't trust the people, don't play with them. The rules certainly don't provide a "balanced atmosphere to prepare for everyone to feel useful" as they stand, without DM involvement. And that's rather on purpose.

[sithlordnergal]

Seeing how most of what players are seeking out of that combination is circumventing opportunity costs of being one or the other Id say multiclassing doesn't serve it's purpose for my games. Easier sure but based on the time players tend to spend with a PC that's a poor excuse to go that route. No reason to rush the foundation.

If a player came to me with a sound reason to mix the two then I might as well take the time to come up with something that works rather than use a janky patchwork of two classes being treated as piles of mechanics.

I mean, I won't deny that you can get people who just multiclass for the abilities, but at the same time I don't think that's a good reason to get rid of multiclassing entirely. You lose a lot of what's fun about DnD by doing so. I also don't really see what you mean by janky patchwork of classes. Classes are mechanics, they're the mechanics you use to uphold your story, no matter what that story is. You can easily make your multiclassing an integrated part of the character's story. You can also take all the class lore and toss it if you want. For example, the Druid class states they revere nature, seek to preserve a balance between nature and civilization, and see themselves as an extension of nature's will.

I see absolutely no reason for a player to be beholden to that lore by any way, shape, or means. I would 100% allow a player to be a Druid that seeks to control and twist nature to their own ends, one that has no concern for that balance, or one that holds a contempt for the natural order of things. I'd suggest to them the Spore or Wildfire subclasses, and tell them to lean into the undead and/or destructive aspects of their subclasses. They're still a Druid because they still have the core Druid abilities, but they're not a regular Druid.

I'll be honest, one of the best things 5e did was remove RP restrictions on classes. No more Lawful Good Paladins, no more Neutral Druids, or Chaotic Bards and Barbarians. The system is better that way because it opens the door for so many other options straight out the gate.



Here's the thing. The system never promises "every single potential option". It says "here are the supported archetypes. Pick one." That's the purpose of a class-level system--to restrict choices to the ones provided.

You want a point-buy system. I don't. D&D is not one. Nor would making it a half-baked point-buy system be to its benefit.


The system doesn't say you're stuck using just those supported archtypes either. Heck, given the fact that they made multiple ways to mix up classes, they're pretty clearly saying "You can play whatever you like" in the system. I'll be honest, I suspect your entire premise of "the purpose of a class-level system is to restrict choices to the ones provided" is inherently flawed. I've yet to find a game that uses a class-level system that does not support multiclassing in some shape or form. Every single one I've run into has a way to mix them.

As for a point-buy system, I'm not a fan of those. They require new players to start with a level of system mastery that simply isn't going to happen. I do agree that D&D is not a point-buy system, and that's to its benefit. But multiclassing doesn't turn it into one, nor is it a half-baked point-buy system as you claim.


EDIT: You know, I suspect we'll just keep going around in circles. I won't convince you that multiclassing is good for DnD, you won't convince me its bad. I find multiclassing to be one of the better elements, and enjoy the fact that optimization allows for characters to take on encounters outside of what they normally can, forcing me to become a lot more creative and tactical with how I build and run encounters, you dislike those sorts of chess-like tactics. That said, I will say this, isn't better to get what we both want? Lets be completely honest, it is far, far easier to simply ban something then it is to create something. We can leave multiclassing in the game, you can say no multiclassing in your games, while I say we allow multiclassing in mine. Then it becomes just as easy for both of us to play the type of game we enjoy without putting extra work on the other.

[Burley]

I'm gonna swoop back in and give my personal reason for multiclassing into Warlock, which isn't just to get Eldritch Blast.

So, my Wizard is in a pirate campaign and, while doing some underwater stuff, he was engulfed by a Shambling Kelp Mound that we needed to exterminate. He was knocked unconscious, got a 20 on the first death save, and woke up inside the monster only to get knocked out again. The group was thoroughly bodied by this encounter and we decided to regroup and try the fight again the next day ( next session).

The exact same thing happened: Got knocked unconscious while being engulfed by the shambling mound, rolled a 20 on the first death save, gasped awake in the kelp-y darkness and got knocked out again-again.

That night, while the rest of the pirate crew was drinking to a lost comrade, my wizard was trying to sleep off four near-deaths in two days. His dreams are flooded with images of seaweed, eels, tentacles and deep, piercing, all-consuming black eyes. "The first was a gift. Now, something is owed."



Multiclassing is a great way to show that characters aren't just classes. I'm not playing Diablo or Golden Axe. My character is a collection of unique ideas, personality and choices. My character's journey is more interesting when he goes off the beaten path and takes the Faustian toll road.

[stoutstien]

I'm gonna swoop back in and give my personal reason for multiclassing into Warlock, which isn't just to get Eldritch Blast.

So, my Wizard is in a pirate campaign and, while doing some underwater stuff, he was engulfed by a Shambling Kelp Mound that we needed to exterminate. He was knocked unconscious, got a 20 on the first death save, and woke up inside the monster only to get knocked out again. The group was thoroughly bodied by this encounter and we decided to regroup and try the fight again the next day ( next session).

The exact same thing happened: Got knocked unconscious while being engulfed by the shambling mound, rolled a 20 on the first death save, gasped awake in the kelp-y darkness and got knocked out again-again.

That night, while the rest of the pirate crew was drinking to a lost comrade, my wizard was trying to sleep off four near-deaths in two days. His dreams are flooded with images of seaweed, eels, tentacles and deep, piercing, all-consuming black eyes. "The first was a gift. Now, something is owed."



Multiclassing is a great way to show that characters aren't just classes. I'm not playing Diablo or Golden Axe. My character is a collection of unique ideas, personality and choices. My character's journey is more interesting when he goes off the beaten path and takes the Faustian toll road.

But multiclassing isn't necessary for organic character development. In a lot of ways it's actively detrimental because of the way level by level multi-classing functions. Not only do you need plan out the prerequisites and need to wait for the "level up" movement for this to occur you now have to actively decide to stop progression as X to start progression as Y.

In my eyes this is a golden opportunity for a player to really have a positive experience shot in the foot by prior edition baggage.
All kinds of boons/blessings/gifts/bargains you could work with before you even take a little bit of prep time to figure out how to make this happenstance shine.

[LibraryOgre]

This right here. Single classes, subclasses, and feats simply can't cover every single potential option for a possible character. You would need an absolutely flood of subclasses to get close to it. That's why I say that just using Subclasses and Feats to emulate multiclassing would do a very, very poor job. You'd end up with a half-assed system where you don't actually get to make the characters you really want to make.

Nor does multiclassing. "This doesn't do everything, so we should use this bad option that does slightly more while introducing more problems" isn't a strong argument.

[PhoenixPhyre]

Nor does multiclassing. "This doesn't do everything, so we should use this bad option that does slightly more while introducing more problems" isn't a strong argument.

This. Very much this

[LibraryOgre]

A simple solution, somewhat coming into vogue? Feat chains.

Magic Initiate lets you pick up a couple spells and cantrips from a given magic class, cast as a member of that magic class. What about a Magic Adept feat, letting you learn a few couple more spells, possibly of higher level? Similar to a bard's Magical Secrets, but requiring more opportunity cost (i.e. a n ASI/feat) and chaining (i.e. you have to have Magic Initiate before Magic Adept)?

Simple. Lets you dabble in things without the kludge that is multiclassing. Takes a couple paragraphs.

[Psyren]

Nor does multiclassing. "This doesn't do everything, so we should use this bad option that does slightly more while introducing more problems" isn't a strong argument.

It does a lot more, not "slightly more." The number of possible builds in a game that bans multiclassing and one that allows it are night and day.


A simple solution, somewhat coming into vogue? Feat chains.

Magic Initiate lets you pick up a couple spells and cantrips from a given magic class, cast as a member of that magic class. What about a Magic Adept feat, letting you learn a few couple more spells, possibly of higher level? Similar to a bard's Magical Secrets, but requiring more opportunity cost (i.e. a n ASI/feat) and chaining (i.e. you have to have Magic Initiate before Magic Adept)?

Simple. Lets you dabble in things without the kludge that is multiclassing. Takes a couple paragraphs.

I'm extremely doubtful that games that consider multiclassing to be too problematic would be on board with a bunch of feat chains getting added to the game instead. In addition, due to how feat acquisition is spaced out in 5e, you're asking people to play something that doesn't actually fit their character concept for multiple levels at a time until they arrive at the ASI "landing" - whereas many multiclass builds let you feel like the thing you're trying to play after just two or three levels.

[Talamare]

It does a lot more, not "slightly more." The number of possible builds in a game that bans multiclassing and one that allows it are night and day.



I'm extremely doubtful that games that consider multiclassing to be too problematic would be on board with a bunch of feat chains getting added to the game instead. In addition, due to how feat acquisition is spaced out in 5e, you're asking people to play something that doesn't actually fit their character concept for multiple levels at a time until they arrive at the ASI "landing" - whereas many multiclass builds let you feel like the thing you're trying to play after just two or three levels.

Press X to Doubt

The number of viable builds allowed is absolutely minor

Sure, it technically allows a million of theoretical possibilities, but most of them are uselessly trash.
It mostly allows for a few exploitative builds that vastly outperform what single class builds can do.

There are VERY few niches that can't currently be fulfilled adequately by an existing subclass, and they aren't even done adding subclasses.

All else fails, literally homebrewing your specific niche into the game into subclass is better than the broken multiclass system.

[stoutstien]

A simple solution, somewhat coming into vogue? Feat chains.

Magic Initiate lets you pick up a couple spells and cantrips from a given magic class, cast as a member of that magic class. What about a Magic Adept feat, letting you learn a few couple more spells, possibly of higher level? Similar to a bard's Magical Secrets, but requiring more opportunity cost (i.e. a n ASI/feat) and chaining (i.e. you have to have Magic Initiate before Magic Adept)?

Simple. Lets you dabble in things without the kludge that is multiclassing. Takes a couple paragraphs.

Eh. I don't like feat chains for the same reason I don't like multiclassing. It might solve one issue by creating a dozen more. I'd rather just put the the entirety of those features into a single feat that evolves over time based on some additional gate like Prof bonus, level, or the like.
All that is assuming you want to keep feats as they are in 5e. If you were to restructure them, which seems the direction they are going, then that's different.still want to avoid 3.x feat chains of Oblivion. Take 4 pointless options to get to the broken one.

[Psyren]

The number of viable builds allowed is absolutely minor

Sure, it technically allows a million of theoretical possibilities, but most of them are uselessly trash.

Most of anything is trash (Sturgeon's Law.) Multiclassing still creates far more good possibilities than banning it does.



There are VERY few niches that can't currently be fulfilled adequately by an existing subclass, and they aren't even done adding subclasses.

Of course single class builds are "adequate." Adequate is a low bar in this game.



All else fails, literally homebrewing your specific niche into the game into subclass is better than the broken multiclass system.

Multiclassing is allowed in AL and at most tables, while homebrew is not.

[stoutstien]

Most of anything is trash (Sturgeon's Law.) Multiclassing still creates far more good possibilities than banning it does.



Of course single class builds are "adequate." Adequate is a low bar in this game.



Multiclassing is allowed in AL and at most tables, while homebrew is not.

I would say 99.999999999% of tables homebrew to some extent even if they don't realize it. trying to maintain a strictly RAW game leads to obscurity.

AL is living proof why multiclassing, and RaW in general taking priority, can lead to epic failure so I don't know why you'd bring that up.

[Segev]

Nor does multiclassing. "This doesn't do everything, so we should use this bad option that does slightly more while introducing more problems" isn't a strong argument.

That's not the argument, though. The argument is, "None of these do everything. All of them being available is a good way to cover more options."

[LibraryOgre]

I'm extremely doubtful that games that consider multiclassing to be too problematic would be on board with a bunch of feat chains getting added to the game instead. In addition, due to how feat acquisition is spaced out in 5e, you're asking people to play something that doesn't actually fit their character concept for multiple levels at a time until they arrive at the ASI "landing" - whereas many multiclass builds let you feel like the thing you're trying to play after just two or three levels.

So, the fact that I consider multiclassing to be problematic, and proposed, as a solution, a feat chain, doesn't dispel your doubt? The fact that I've advocated feats as a way to do away with TAL multiclassing since the beginning doesn't dispel your doubt?

As for "Well, you can do it in two or three levels if you're multiclassed", you can also do that if you're single classed and use a subclass. "I am a fighter who is also a wizard." You mean Eldritch Knight, which comes on at 3rd level? "I study necromancy of all kinds?" Oh, so you're a grave any cleric with the Arcane skill, who will pick up feats as you're able to master more complex magics?

Seriously, I can manage the "Feat Chain" with one feat.

Magical Adept
Prerequisite: Magic Initiate
You have deepened and expanded your magical knowledge and abilities. Choose two new spells from the spell list of a class you have chosen for Magic Initiate. You learn two spells from that class spell list. You may substitute one of those spells for two cantrips. These spells may be of 2nd or higher level, but you must possess one spell of each successive level; you may not learn a 3rd level spell until you have learned a 2nd level spell from a given class, nor a 5th level spell until you have learned a 4th level spell. As with Magic Initiate, these spells may either be cast once per long rest at their lowest level, or with your own spell slots, and continue to use the spellcasting ability of the chosen class. Spells selected through this feat may not exceed 5th level. This feat may be taken multiple times.

Boom. I don't want to play an Arcane Trickster? As a Thief, I can now cast 2 cantrips and a 1st level spell at 4th, a 2nd and 3rd at 8th, and a 4th and a 5th at 10th (and if the Arcane Trickster wants to do this, they can, too, and they have spell slots to be better at it than the Thief). Do I want to be a cleric with a bit of wizard? Those come on-line at 4th (1st level), 8th (up to 3rd), and 12th (up to 5th), and can power these with the spell slots I have. I might slow down a bit if I also want Ritual Caster.

[Segev]

So, the fact that I consider multiclassing to be problematic, and proposed, as a solution, a feat chain, doesn't dispel your doubt? The fact that I've advocated feats as a way to do away with TAL multiclassing since the beginning doesn't dispel your doubt?

As for "Well, you can do it in two or three levels if you're multiclassed", you can also do that if you're single classed and use a subclass. "I am a fighter who is also a wizard." You mean Eldritch Knight, which comes on at 3rd level? "I study necromancy of all kinds?" Oh, so you're a grave any cleric with the Arcane skill, who will pick up feats as you're able to master more complex magics?

Seriously, I can manage the "Feat Chain" with one feat.

Magical Adept
Prerequisite: Magic Initiate
You have deepened and expanded your magical knowledge and abilities. Choose two new spells from the spell list of a class you have chosen for Magic Initiate. You learn two spells from that class spell list. You may substitute one of those spells for two cantrips. These spells may be of 2nd or higher level, but you must possess one spell of each successive level; you may not learn a 3rd level spell until you have learned a 2nd level spell from a given class, nor a 5th level spell until you have learned a 4th level spell. As with Magic Initiate, these spells may either be cast once per long rest at their lowest level, or with your own spell slots, and continue to use the spellcasting ability of the chosen class. Spells selected through this feat may not exceed 5th level. This feat may be taken multiple times.

Boom. I don't want to play an Arcane Trickster? As a Thief, I can now cast 2 cantrips and a 1st level spell at 4th, a 2nd and 3rd at 8th, and a 4th and a 5th at 10th (and if the Arcane Trickster wants to do this, they can, too, and they have spell slots to be better at it than the Thief). Do I want to be a cleric with a bit of wizard? Those come on-line at 4th (1st level), 8th (up to 3rd), and 12th (up to 5th), and can power these with the spell slots I have. I might slow down a bit if I also want Ritual Caster.
More that I am opposed to feat chains, myself, in 5e. It's just that 5e's feats are too "big," granularly speaking, and come too infrequently.

That said, as an additional option? Sure, why not? I don't want chains, though; I'd prefer singletons. If we MUST chain, so be it, but...

Heck. Feats are, in general, MORE VALUABLE than class levels in 5e.

[Psyren]

I would say 99.999999999% of tables homebrew to some extent even if they don't realize it. trying to maintain a strictly RAW game leads to obscurity.

AL is living proof why multiclassing, and RaW in general taking priority, can lead to epic failure so I don't know why you'd bring that up.

AL has failed? Someone should have announced that to the two-dozen-plus full tables I saw playing it at DragonCon this past weekend then.


So, the fact that I consider multiclassing to be problematic, and proposed, as a solution, a feat chain, doesn't dispel your doubt? The fact that I've advocated feats as a way to do away with TAL multiclassing since the beginning doesn't dispel your doubt?

I have no doubts that you're on board with this solution, no. And I'm not saying it lacks merit either.



As for "Well, you can do it in two or three levels if you're multiclassed", you can also do that if you're single classed and use a subclass. "I am a fighter who is also a wizard." You mean Eldritch Knight, which comes on at 3rd level? "I study necromancy of all kinds?" Oh, so you're a grave any cleric with the Arcane skill, who will pick up feats as you're able to master more complex magics?

There are certainly subclasses that enable hybrid concepts, I'm not denying that those exist. But not all of them do. There is not yet a paladin subclass that makes them excel at ranged combat for instance, but there are multiclass builds that do. There is not yet a barbarian subclass that allows them to grapple any monster in the game, but there are multiclass builds that do. There is not yet a rogue that can hide in the darkness without using magic from creatures with darkvision, but there are multiclass builds that do. And so on.

[Captain Cap]

Besides, subclasses are exactly that, subclasses: they don't carry the same weight of your primary class and so would never be able to produce an "exact" hybrid, unless you revise completely the concept.

[LibraryOgre]

There are certainly subclasses that enable hybrid concepts, I'm not denying that those exist. But not all of them do. There is not yet a paladin subclass that makes them excel at ranged combat for instance, but there are multiclass builds that do. There is not yet a barbarian subclass that allows them to grapple any monster in the game, but there are multiclass builds that do. There is not yet a rogue that can hide in the darkness without using magic from creatures with darkvision, but there are multiclass builds that do. And so on.

And there are multiclass builds that let you turn warlock spell slots into tons of sorcery points. And there are multiclass builds that let you shoot an Eldritch Blast 1200 feet. Or learn every skill in the game. Or be a wildshaped druid with resistance to everything except psychic damage.

[Psyren]

And there are multiclass builds that let you turn warlock spell slots into tons of sorcery points. And there are multiclass builds that let you shoot an Eldritch Blast 1200 feet. Or learn every skill in the game. Or be a wildshaped druid with resistance to everything except psychic damage.

I only see the first one of these as close to being a problem, and that one is easily fixed without removing the system entirely.

(...Out of curiosity, what's the EB one?)

[JNAProductions]

I only see the first one of these as close to being a problem, and that one is easily fixed without removing the system entirely.

(...Out of curiosity, what's the EB one?)

Eldritch Spear, Spell Sniper, Distant Spell.

[Psyren]

Eldritch Spear, Spell Sniper, Distant Spell.

Eh, that's an issue with allowing things to be doubled multiple times, not with multiclassing. Especially since you can do this combo on a single-classed Warlock.

[stoutstien]

AL has failed? Someone should have announced that to the two-dozen-plus full tables I saw playing it at DragonCon this past weekend then.
.

I didn't say AL was a failure I said it leads to failure. If you try to use it as a point of comparison it disregards most of what TTRPGs strive to achieve. Con play is about the only place it makes any sense because it's they are generally one shot in nature and don't need to worry about stuff like pacing or consistency.

IMO it's not even the same game. It's a loading que for tables or purgatory for those who can't find one.

[Psyren]

I didn't say AL was a failure I said it leads to failure. If you try to use it as a point of comparison it disregards most of what TTRPGs strive to achieve. Con play is about the only place it makes any sense because it's they are generally one shot in nature and don't need to worry about stuff like pacing or consistency.

IMO it's not even the same game. It's a loading que for tables or purgatory for those who can't find one.

This all reeks of disdain for organized play that suffice to say, I don't think WotC shares in the slightest.

[stoutstien]

This all reeks of disdain for organized play that suffice to say, I don't think WotC shares in the slightest.

Seeing the usual pool of players at the local I actually love AL. It's the best buffer created for DMs to form tables that work for them rather than clobbering together mismatched expectations.

[Rukelnikov]

I'm gonna swoop back in and give my personal reason for multiclassing into Warlock, which isn't just to get Eldritch Blast.

So, my Wizard is in a pirate campaign and, while doing some underwater stuff, he was engulfed by a Shambling Kelp Mound that we needed to exterminate. He was knocked unconscious, got a 20 on the first death save, and woke up inside the monster only to get knocked out again. The group was thoroughly bodied by this encounter and we decided to regroup and try the fight again the next day ( next session).

The exact same thing happened: Got knocked unconscious while being engulfed by the shambling mound, rolled a 20 on the first death save, gasped awake in the kelp-y darkness and got knocked out again-again.

That night, while the rest of the pirate crew was drinking to a lost comrade, my wizard was trying to sleep off four near-deaths in two days. His dreams are flooded with images of seaweed, eels, tentacles and deep, piercing, all-consuming black eyes. "The first was a gift. Now, something is owed."



Multiclassing is a great way to show that characters aren't just classes. I'm not playing Diablo or Golden Axe. My character is a collection of unique ideas, personality and choices. My character's journey is more interesting when he goes off the beaten path and takes the Faustian toll road.

This is by far the most important reason why choosing what you take on every level up is the better path, it makes the characters more organic, you can choose your level up reflecting what happened to you since the previous level up, usually your gonna keep in your "path", but events make you deviate from it, otherwise you are picking a train at chargen and can only go where the tracks take you.


Nor does multiclassing. "This doesn't do everything, so we should use this bad option that does slightly more while introducing more problems" isn't a strong argument.

I think this is a major contention point, IMO multiclassing adds more than what is left without it.


A simple solution, somewhat coming into vogue? Feat chains.

Magic Initiate lets you pick up a couple spells and cantrips from a given magic class, cast as a member of that magic class. What about a Magic Adept feat, letting you learn a few couple more spells, possibly of higher level? Similar to a bard's Magical Secrets, but requiring more opportunity cost (i.e. a n ASI/feat) and chaining (i.e. you have to have Magic Initiate before Magic Adept)?

Simple. Lets you dabble in things without the kludge that is multiclassing. Takes a couple paragraphs.

Dropping the classes almost entirely and replacing them for something like this is what 3e Unearthed Arcana "classless" system proposed, I always hoped they went that route and further eliminated classes entirely, the latest 5e UAs prior to DnD1 reveal gave me a little hope that they may be considering that direction more.

[sithlordnergal]

Nor does multiclassing. "This doesn't do everything, so we should use this bad option that does slightly more while introducing more problems" isn't a strong argument.

Better to have it than not, and I also feel its actually a really good system. It also doesn't introduce any more problems from what I've seen and played. At least, no more than feats or magic items do. My only complaint is that they probably should lessen the multiclass restrictions on ASIs in the next version.



But multiclassing isn't necessary for organic character development. In a lot of ways it's actively detrimental because of the way level by level multi-classing functions. Not only do you need plan out the prerequisites and need to wait for the "level up" movement for this to occur you now have to actively decide to stop progression as X to start progression as Y.

In my eyes this is a golden opportunity for a player to really have a positive experience shot in the foot by prior edition baggage.
All kinds of boons/blessings/gifts/bargains you could work with before you even take a little bit of prep time to figure out how to make this happenstance shine.

Multiclassing may not be necessary for organic character development, but it sure does help uphold that character development and show how that development has affected the character. And I fully disagree that its detrimental. You do need to worry about pre-reqs, but those were a dumb restriction anyway, and one they'll hopefully remove or lessen in the next edition while keeping multiclassing level-by-level. Everything else is completely natural. After all, you need to wit for a level up moment anyway before you can decide to continue your progression in your current class, and nothing prevents you from deciding to RP researching the new class you'll be going into between the level ups.

For my own little anecdote, my first character I ever played was an Ancients Paladin. We were at level 7 when we began dealing with a lot of wild magic from the Far Realm. My DM and I both agreed that this should have some effect on me since I'm basically exuding an aura that protects my allies from the effects of magic, via halving spell damage and granting a bonus to saves. Instead of coming up with some half-assed homebrew that wouldn't have been very good, we both figured multiclassing into Wild Magic Sorcerer would be a far better option. And I did. And is where I got my love for the Wild Magic Sorcerer. No need to give some boon that gives me access to the wild magic table, just a level of Wild Magic Sorcerer.



Press X to Doubt

The number of viable builds allowed is absolutely minor

Sure, it technically allows a million of theoretical possibilities, but most of them are uselessly trash.
It mostly allows for a few exploitative builds that vastly outperform what single class builds can do.

There are VERY few niches that can't currently be fulfilled adequately by an existing subclass, and they aren't even done adding subclasses.

All else fails, literally homebrewing your specific niche into the game into subclass is better than the broken multiclass system.

You shouldn't doubt. Consider just the Fighter and Rogue, PHB subclasses alone. From those 6 options, you have 18 combinations, each one perfectly viable, and each one plays differently. Even if you treat things like Eldritch Knight/Thief and Thief/Eldritch Knight combinations the same, which you shouldn't cause they have different saves, skills, equipment and actually play differently, you're still looking at 9 different combinations of Subclasses. And within those 9, you have a multitude of variety depending on how many levels you take of each class.

As for "viable builds" in general, I have only seen one build that wasn't "viable", and that was Puffin Forest's Abserd build where he only took 1 or 2 levels of every single class in the game. Everything else? Viable. Monk/Barbarian? Viable, even with 27 Point Buy and restrictions on multiclassing ability scores. Barbarian/Wizard? 100% viable, and rather fun to play I might add.

And I disagree with there being few niches that can't be fulfilled adequately via subclass alone. My Ancients Paladin/Dream Druid does a far better job at being a protector of the land and bulwark against the undead tide in Chult then a single classed Paladin or Druid would, and infinitely better than a Nature Cleric ever could. A Fighter/Wizard does a far better job of being a war wizard than War Wizards or Eldritch Knights do. Barbarian/Rogue is far better at depicting the brutish muscle of the thieves' guild than a Barbarian or Rogue do on their own.

And finally, why bother homebrewing if multiclassing exists? Hell, as a DM I would far prefer a player multiclass than create an entirely new subclass on the fly, make sure those subclass abilities are properly balanced with everything else and the class being chosen, on top of making the entire world, all the encounters, and then some. If you want to homebrew, fine, be my guest, but leave multiclassing in because it makes my life as a DM infinitely easier without taking away a player's ability to fill in whatever niche they want.

[Psyren]

And it's not like allowing multiclassing prevents people from homebrewing in any event.

[PhoenixPhyre]

And it's not like allowing multiclassing prevents people from homebrewing in any event.

It does make homebrewing way more complicated in my experience. Unless you outright lock out any homebrew from being used in a multiclassing environment. Because now you have to consider all possible combinations of things. And since you can't really play-test any of that (playtesting homebrew is hard enough, let alone combinatorial explosion of possibilities), it's really easy for wacky things to slip by.

And it also provides negative pressure on homebrew--"just multiclass instead" is the "easy" way out. Despite it usually not fitting the world or the character or the campaign nearly as well (isn't it convenient how all those wizards suddenly find religion and get chosen by gods...but just long enough. Not long enough to actually continue on as clerics, not forsaking their earlier path for more than a few days, just "here, have some nice goodies, now never worry about the gods again.").

[Psyren]

I don't agree that you need a bunch more cleric levels to show your willingness to "continue on the path." A wizard/cleric who is mostly wizard, would probably still be an appealing and cherished follower to someone like Azuth, Mystra, Boccob or Nethys.

I do agree that homebrew + multiclassing is definitely more complicated. They're also both optional (homebrewing exponentially so) so presumably tables allowing both think they can handle it. And if it turns out to be more disruptive than expected, just have a conversation with the player and come to an agreement around changes to their character.

[sithlordnergal]

It does make homebrewing way more complicated in my experience. Unless you outright lock out any homebrew from being used in a multiclassing environment. Because now you have to consider all possible combinations of things. And since you can't really play-test any of that (playtesting homebrew is hard enough, let alone combinatorial explosion of possibilities), it's really easy for wacky things to slip by.

I mean, that's just a general Homebrewing issue, not a multiclass one. Homebrewing classes and subclasses will always have that risk, be it with feats, races, or skills. Its why I tend to dislike homebrew in my games. I'm busying building fun and challenging encounters and puzzles that make my players think strategically, while also making the world and story. I don't have time to build an entire class and subclass. And the easy solution to that is "No multiclassing with homebrew" or "Restricted multiclassing with homebrew". And if you're playtesting a class, you should probably inform a player of that fact, that way they can take the notes you need to fix up any unexpected bugs the class might have.

No need to strip out a good and useful system in dnd just because you use homebrew.



And it also provides negative pressure on homebrew--"just multiclass instead" is the "easy" way out. Despite it usually not fitting the world or the character or the campaign nearly as well (isn't it convenient how all those wizards suddenly find religion and get chosen by gods...but just long enough. Not long enough to actually continue on as clerics, not forsaking their earlier path for more than a few days, just "here, have some nice goodies, now never worry about the gods again.").

I don't really see how it provides "negative pressure" to homebrew. It does make certain options easier, and the lives of DMs easier though. I run 3 to 4 games a week, with seperate tables, in seperate worlds that don't have any connection to each other. I just don't have time to build an entire new class with subclasses just to replicate a Fighter/Rogue, Paladin/Sorcerer, or Wizard/Cleric. And I don't have time to look over homebrew classes other people have made in order to find their strengths and weaknesses. If a player brings me a homebrew class, I'll look over it and determine if its ok, but I don't have time to do all that myself.

And that just sounds like a poor roleplaying issue. Something that is not a multiclass issue.

[Psyren]

A more lightweight way to "homebrew" (and to multiclass for that matter) is to simply take a weak subclass feature out of one subclass and swap it with a stronger one - ideally from the same class, but possibly from a different one. For example, one of our DMs took Infiltration Expertise and Impostor out of the Assassin, and swapped them for Superior Mobility and Ambush Master from the Scout. The end result still felt like an assassin and was still very uncomplicated to run at the table.

There are even guidelines for this process on DMG 287.

[PhoenixPhyre]

I don't agree that you need a bunch more cleric levels to show your willingness to "continue on the path." A wizard/cleric who is mostly wizard, would probably still be an appealing and cherished follower to someone like Azuth, Mystra, Boccob or Nethys.

I do agree that homebrew + multiclassing is definitely more complicated. They're also both optional (homebrewing exponentially so) so presumably tables allowing both think they can handle it. And if it turns out to be more disruptive than expected, just have a conversation with the player and come to an agreement around changes to their character.

But "appealing and cherished follower" =/= "living hand of the god". Which is what a cleric is. The whole point of being a cleric is that you're chosen to be the god's mortal instrument, forsaking your previous life. It's not "you're a nice guy, here's some power."

And homebrew isn't optional--it's actually mandatory. Because any time you make rulings? You're homebrewing. Homebrew is expected, which is why there's all those pieces of the DMG devoted to it. And it affects more than just classes--everything has to now be balanced against the entire set of possibilities. Not just one. So yes, multiclassing makes homebrewing (which is much more fundamental to TTRPGs than level-by-level multiclassing, a thing basically only known in 3e and 5e D&D) much harder. For, IMO, little benefit beyond "I can make more powerful characters."


I mean, that's just a general Homebrewing issue, not a multiclass one. Homebrewing classes and subclasses will always have that risk, be it with feats, races, or skills. Its why I tend to dislike homebrew in my games. I'm busying building fun and challenging encounters and puzzles that make my players think strategically, while also making the world and story. I don't have time to build an entire class and subclass. And the easy solution to that is "No multiclassing with homebrew" or "Restricted multiclassing with homebrew". And if you're playtesting a class, you should probably inform a player of that fact, that way they can take the notes you need to fix up any unexpected bugs the class might have.

No need to strip out a good and useful system in dnd just because you use homebrew.


"No multiclassing with homebrew" === "stripping out a 'good and useful' (scare quotes intentional) system in dnd just because you use homebrew". There's no two ways about it--multiclassing makes homebrew harder. And homebrew provides tons of value. In fact, it provides the only value to a TTRPG over a computerized one--the ability to insert your own content. And yes, this includes all homebrew, not just full classes or even subclasses. Feats, items, monsters, spells, etc. All have to now be validated against the entire world, not just a single element.



I don't really see how it provides "negative pressure" to homebrew. It does make certain options easier, and the lives of DMs easier though. I run 3 to 4 games a week, with seperate tables, in seperate worlds that don't have any connection to each other. I just don't have time to build an entire new class with subclasses just to replicate a Fighter/Rogue, Paladin/Sorcerer, or Wizard/Cleric. And I don't have time to look over homebrew classes other people have made in order to find their strengths and weaknesses. If a player brings me a homebrew class, I'll look over it and determine if its ok, but I don't have time to do all that myself.

And that just sounds like a poor roleplaying issue. Something that is not a multiclass issue.

It's a roleplay issue made possible and even worse by multiclassing. It's something (archetype breakage) that can't really come up without it. Yet is the norm with it. Seriously, level-by-level multiclassing relies on the model of "classes are just grab-bags of mechanics, fluff is optional" to make most of the combinations plausible. It also stretches belief--a normal person has to work for a decade to become a first level wizard. Yet you picked it up, while adventuring, over the course of a single day. Or you just picked up a Patron and Pact...without any event in character. Overnight, while you were asleep after a long day of paladining. Yes, you can work around it. But now the whole campaign and world have to adapt to that, for what are dominantly mechanical reasons. And every time the mechanics force the fiction to bend (rather than the other way around), that's a sign something's gone wrong.

[JNAProductions]

You should be inspired by the classes’ lore.
Not restrained by it.

[LibraryOgre]

I mean, that's just a general Homebrewing issue, not a multiclass one.

But that's been the point... it IS the multiclass issue. Exactly the multiclass issue we've been talking about, that you dismiss as unimportant. Without the TAL multiclass, a well-built homebrew doesn't run into those problems any more than a standard class, but with TAL multiclass, you wind up with "Well, that was unexpected". Yes, you can have homebrew that creates monsters on purpose, or does so by accident, but THAT is the fault of homebrew. With TAL multiclass, you don't only need to consider your mechanics, but all core mechanics... your novel mechanic that works fine in its own context creates an unintentional monster when combined with something else, where it would not if you didn't have TAL multiclass.

[stoutstien]

A more lightweight way to "homebrew" (and to multiclass for that matter) is to simply take a weak subclass feature out of one subclass and swap it with a stronger one - ideally from the same class, but possibly from a different one. For example, one of our DMs took Infiltration Expertise and Impostor out of the Assassin, and swapped them for Superior Mobility and Ambush Master from the Scout. The end result still felt like an assassin and was still very uncomplicated to run at the table.

There are even guidelines for this process on DMG 287.

The DMG base function is a primer on how to homebrew because it's a necessary component of the game. Call it DM ruling, fiat, adjudication, house rules, homebrew, or whatever you like but it's one of the basic components you need.

[PhoenixPhyre]

You should be inspired by the classes’ lore.
Not restrained by it.

Disagree, somewhat. The whole point of having classes is archetype enforcement. And for that to happen,
* classes need to fit/tie into archetypes (even if those are synthetic things nowhere seen before)--classes need to have "lore" (class fiction).
* that class fiction needs to differentiate "is part of class X" from "is part of class Y". Which means it needs to be at least partially binding.
* the class mechanics need to uphold that class fiction

Classes that are just "bags of mechanics" aren't actually serving the purpose of having classes in the first place. But the class fiction can depend on the world and the table, and exactly how tightly you hold to it is up to the taste of the table. But it certainly should provide pressure. This is why both fighter and wizard (especially wizard) are such poorly-designed classes--they have no meaningful class fiction encoded in the mechanics.

Edit @Psyren & @sithlordnergal -- homebrew that gets hurt by multiclassing includes encounters. Which are mandatory homebrew unless you're playing a module exactly by the book. Which sucks and misses the entire point of a TTRPG. So your beloved "chess match" encounters? Way harder and more prone to unexpected breakage with multiclassing. Exponentially so--the possibility field increases combinatorially.

[Talamare]

You shouldn't doubt. Consider just the Fighter and Rogue, PHB subclasses alone. From those 6 options, you have 18 combinations, each one perfectly viable, and each one plays differently. Even if you treat things like Eldritch Knight/Thief and Thief/Eldritch Knight combinations the same, which you shouldn't cause they have different saves, skills, equipment and actually play differently, you're still looking at 9 different combinations of Subclasses. And within those 9, you have a multitude of variety depending on how many levels you take of each class.

As for "viable builds" in general, I have only seen one build that wasn't "viable", and that was Puffin Forest's Abserd build where he only took 1 or 2 levels of every single class in the game. Everything else? Viable. Monk/Barbarian? Viable, even with 27 Point Buy and restrictions on multiclassing ability scores. Barbarian/Wizard? 100% viable, and rather fun to play I might add.

And I disagree with there being few niches that can't be fulfilled adequately via subclass alone. My Ancients Paladin/Dream Druid does a far better job at being a protector of the land and bulwark against the undead tide in Chult then a single classed Paladin or Druid would, and infinitely better than a Nature Cleric ever could. A Fighter/Wizard does a far better job of being a war wizard than War Wizards or Eldritch Knights do. Barbarian/Rogue is far better at depicting the brutish muscle of the thieves' guild than a Barbarian or Rogue do on their own.

And finally, why bother homebrewing if multiclassing exists? Hell, as a DM I would far prefer a player multiclass than create an entirely new subclass on the fly, make sure those subclass abilities are properly balanced with everything else and the class being chosen, on top of making the entire world, all the encounters, and then some. If you want to homebrew, fine, be my guest, but leave multiclassing in because it makes my life as a DM infinitely easier without taking away a player's ability to fill in whatever niche they want.

Viable means that its at least equal to someone who didn't multiclass. Claiming that every potential multiclass is at minimally equal to a single class is either wrong or part of the problem.

The statement being wrong means that not every multiclass is equal to a single class, which honestly is understandable...
or its a problem because that means multiclasses are generally inherently extremely powerful...

Probably because you know...
They frontloaded a lot of features because they expect that the game won't really see the higher play levels.

So, when you go around maximizing those powerful early features you tend to end up with builds that are far beyond a generic single classer.


Honestly, if you trim down Abserd slightly by cutting out only a few classes you would have an Absurdly powerful build. (Okay, it's not Absurdly powerful, but you legit can make strong builds with several multiclasses which actually is Absurd.) ... heh


I honestly hate homebrewing overall. I rather people fulfill their fantasy without being so mentally restricted on the class.
I've played a Travelling Merchant who went around stealing stuff then reselling, guess what class I played? Bard? Rogue? Nope... Druid, I changed the fluff on a lot of stuff, but no mechanics and it made sense.

What I hate even more than homebrewing are people who bring absolutely brokenly powerful multiclass heavy builds that no single class can achieve as they start explaining how technically the wording on certain things basically says it should work.

[Witty Username]

If one of the problems with multiclassing is features needing to be balanced with every class, wouldn't feats have a similar issue?

Especially feat chains that have a similar collection of features, that classes do?

[stoutstien]

If one of the problems with multiclassing is features needing to be balanced with every class, wouldn't feats have a similar issue?

Especially feat chains that have a similar collection of features, that classes do?

feats have a similar issue to some degree but because they have less potential combinations and points of access to consider they are easier to fold in.
It's a basic principle that if you add things that can be stacked rather than selected the number of possible combinations are greatly increased (which is the major point those who like it tend to fall back on as a defense). However the odds of unintentional interactions also increases at a similar rate.

If you were introducing a feat in the game and you want to see how it interact with each class you could do that without much gripe but if you wanted to see how just 2 classes multiclassing impact each other you have a ton of combos and a ton of timings considerations to contend with. When features are added is as, if not more important, that what features are added.

[Talamare]

If one of the problems with multiclassing is features needing to be balanced with every class, wouldn't feats have a similar issue?

Especially feat chains that have a similar collection of features, that classes do?

Absolutely, and we assume they take the time to balance them against every class/combo

If anything, removing multiclass makes it easier to balance feats, since now you just need to worry about every subclass instead of every possible subclass combination...

[LibraryOgre]

If one of the problems with multiclassing is features needing to be balanced with every class, wouldn't feats have a similar issue?

Especially feat chains that have a similar collection of features, that classes do?

To a lesser extent, but yes. A big part of it is that feats are more limited... they can have an unbalancing effect, but with fewer moving parts, they can't do as much system damage.

Consider, for a moment, the Big Two of martial feats... Great Weapon Master and Sharpshooter. These feats are very powerful, arguably broken, and one way to tell that is that they figure in almost every build that emphasizes melee or ranged combat, respectively. Breaking them each into two feats (bonus attack and damage, or range & cover and damage) might wind up with them being too weak, but as they stand, an indication of their strength is how prominently they figure into many builds. Martial Adept, giving you a limited version of the Battlemaster subclass maneuvers? Not nearly as critical in builds, because they're more limited.

And that's where feats become less of an issue over multiclassing: feats are a lot more limited.1 They don't have the versatility of a class feature. A Coffelock's sorcery points/short rest warlock spells, though, provides an incredible amount of strength, with extreme versatility. And that's where things break... versatility mixed with disparate mechanics. A coffelock is completely negated if you put a warlock on the standard long rest slot allocation... the power comes from the interaction between short rest spell slots and converting those slots to spell points. A 2nd level sorcerer/3rd level warlock has as many sorcery points across three short rests as a 20th level sorcerer (2 from being a sorcerer, 6 per short rest from being a Warlock). At 3/3, that coffeelock has 21 sorcery points over those 3 short rests, or 9 without a short rest. Once they hit that, they also have metamagic, and their plethora of sorcery points means they can ramp up their spellcasting even further. That synergy is a versatile one. (It gets even worse if the coffelock is an elf... long rest in 4 hours, then take short rests to charge up sorcery points, while the rest of the party finishes their long rest)

Metamagic Adept, from Tasha's Cauldron of Everything? 2 Sorcery Points and 2 known metamagics? Still very useful to a pure warlock, but the lack of ability to convert spell slots into sorcery points means that it does not have the synergy you get from being a sorcerer/warlock. Arguably, it is also a lot less of an impact to say "You can't take this feat" than it is to say "We allow multiclassing, but not these certain combinations", especially if your argument in favor of multiclassing is that it lets you make character concepts that you otherwise could not.

TL;DR: Feats are more limited, and so have less of an overall impact than combining class features.

1 While it thankfully hasn't come up here, I will emphasize... we're talking about well and reasonably constructed things, here. You can certainly make a class or a feat that breaks the game in and of itself, but that's not the issue when discussing the problems of synergy

[Psyren]

But "appealing and cherished follower" =/= "living hand of the god". Which is what a cleric is. The whole point of being a cleric is that you're chosen to be the god's mortal instrument, forsaking your previous life. It's not "you're a nice guy, here's some power."

If a god can in no way make use of a "mortal instrument" who is more adept with arcane magic than divine, I don't think that god is very creative, or good at their job. Especially a god of magic, which would be the most likely sphere to have such a multiclass among their servants.



And homebrew isn't optional--it's actually mandatory. Because any time you make rulings? You're homebrewing. Homebrew is expected, which is why there's all those pieces of the DMG devoted to it. And it affects more than just classes--everything has to now be balanced against the entire set of possibilities. Not just one. So yes, multiclassing makes homebrewing (which is much more fundamental to TTRPGs than level-by-level multiclassing, a thing basically only known in 3e and 5e D&D) much harder. For, IMO, little benefit beyond "I can make more powerful characters."

"No multiclassing with homebrew" === "stripping out a 'good and useful' (scare quotes intentional) system in dnd just because you use homebrew". There's no two ways about it--multiclassing makes homebrew harder. And homebrew provides tons of value. In fact, it provides the only value to a TTRPG over a computerized one--the ability to insert your own content. And yes, this includes all homebrew, not just full classes or even subclasses. Feats, items, monsters, spells, etc. All have to now be validated against the entire world, not just a single element.

I agree with you that rulings are necessary for a TTRPG like 5e to work, but I don't agree with you that rulings and homebrew can be considered as one and the same thing. The homebrew board on this forum is not dedicated to everything that could be considered a ruling, there's a clear difference.

[LibraryOgre]

I agree with you that rulings are necessary for a TTRPG like 5e to work, but I don't agree with you that rulings and homebrew can be considered as one and the same thing. The homebrew board on this forum is not dedicated to everything that could be considered a ruling, there's a clear difference.

I'll also note that I don't consider homebrewing, houseruling, or rulings to be the same thing.

Homebrewing creates a thing... race, class, feat, monster.
Houseruling creates an alteration to an understood process... "Every spellcaster gets the Wizard's Arcane Recovery feature, wizards get it twice per long rest".
Rulings clarify something that is unclear or questionable... "Does my proficiency in brewing supplies let me add my proficiency bonus to appraise the value of this wine?" Not 100% clear, a yes or no could be justified, but a DM call to make (and hopefully be consistent with).

I also don't think any of them are particularly relevant to this conversation (as it has evolved), aside as a deflection from the question of multiclassing. You can houserule things away, and it doesn't change the fact of the design. You can make a ruling about wording, but that can verge on houserules when the wording isn't very clear. And you can homebrew anything, and may have to in order to cover where there's not an existing subclass to cover a combination, but that doesn't mean that homebrew is a priori bad or good, and doesn't change the difficulties highlighted about multiclassing.

[LibraryOgre]

The Mod Ogre: Thinking about it, let's move the discussion of multiclassing v. subclass and feats to a new thread; we're far afield from the original question. If you want to move quotes from here to the new thread to keep continuity, feel free.