Multiclass. Revamp to stop optimization?

[Whit]

1. Is multiclassing for 1 lvl dip or even 2,3 lvls optimizing? Or combo optimizing and backstory?
2. Should multiclassing be revamped to stop the 1-3 lvl dip?
3. If so, how should it be changed? No less than 1-2 levels below others?
4. Increase ability scores to multi class?
5. Which classes are mostly dipped for 1 lvl

[Maxilian]

Well is true that many dips are for the sake of optimization, but in many cases those optimization is just to optimize a playstyle and not to create an OP character, so is rarely a problem imho.

Note: Ability score requirement increase will just discourage Multiclass in general

[Whit]

True Max. But is there really an OP build anymore? It’s usually optimizing now.

[Eragon123]

If dipping bothers you just disallow multiclassing from your game.

There is nothing stopping a sorcerer from swearing an oath.
A wizard from selling their soul.
A barbarian seeking inner peace.

In the end giving out minor rewards will work 90% of the time and is much easier than dealing with multiclassing. especially if your players are prone to cheesing it.

[Maxilian]

True Max. But is there really an OP build anymore? It’s usually optimizing now.

Well... yeah, but we call it to "chesse" it (Like Coffee-locks getting an absurb amount of spell slots if given enough preparation time)

[Maxilian]

In the end giving out minor rewards will work 90% of the time and is much easier than dealing with multiclassing. especially if your players are prone to cheesing it.

I don't really agree, as one can easily go out of hand while the other is harder for it to get out of hand (Giving minor rewards, will most likely means giving homebrew rewards, so its way easier for it to get out of hands than the MC per se -as for the minor rewards, do not cost anything but are in the hands of the DM for when to give)

[Theodoxus]

I used ZMan's homebrew and turned multiclassing into feats. It's worked really well since there's a tax associated with it; not everyone wants to go down that road.

[Maxilian]

Maybe the best way is to have it introduced in the story, have MC happen in the actual story and not just "I'm a Seeker (Random MC name)", but don't over complicate it, or it will just be the same as not allowing MC in general (have the PC do some simple things, like start using some of its time in Short rest to pray, or study a spell book that they found not long ago -or write their own spell book-, practice with the magic of the sound, etc...)

[Eragon123]

I don't really agree, as one can easily go out of hand while the other is harder for it to get out of hand (Giving minor rewards, will most likely means giving homebrew rewards, so its way easier for it to get out of hands than the MC per se -as for the minor rewards, do not cost anything but are in the hands of the DM for when to give)

That's fair. My ideas of minor rewards were mostly just items.

Sorcerer takes an oath to protect the innocent? Maybe a shortsword with a war ruby attached so he can use it as a spellcasting focus.

Wizard who sells his soul? Item that allows him to cast a certain spell a few times per day.

Barbarian seeking inner peace, maybe a small squirrel friend who helps calm him down.

[Maxilian]

That's fair. My ideas of minor rewards were mostly just items.

Sorcerer takes an oath to protect the innocent? Maybe a shortsword with a war ruby attached so he can use it as a spellcasting focus.

Wizard who sells his soul? Item that allows him to cast a certain spell a few times per day.

Barbarian seeking inner peace, maybe a small squirrel friend who helps calm him down.

Would easily bring the same problem.

As items balance is not as easy as the game wish it to be (I mean... not all uncommon items are equally good -same with all types of magic items)

Note: But i do agree with you that it is a great idea to give some character more flavor, without them having to MC or make a big change in their character that they may regret in the future.

[Eragon123]

Would easily bring the same problem.

As items balance is not as easy as the game wish it to be (I mean... not all uncommon items are equally good -same with all types of magic items)

Note: But i do agree with you that it is a great idea to give some character more flavor, without them having to MC or make a big change in their character that they may regret in the future.

At the end of the day there are MANY ways for things to break and that potential can come from any source. Just maintaining dialogue with your players will solve many of these. So while no one solution is perfect, its best to for everyone to present what they do so new ideas can form or even old ideas adopted and spread.

[Grod_The_Giant]

1. Is multiclassing for 1 lvl dip or even 2,3 lvls optimizing? Or combo optimizing and backstory?
Yes. Both reasons are valid. Sometimes even for the same character.


2. Should multiclassing be revamped to stop the 1-3 lvl dip?
No. There's already a pretty significant cost in delayed access to your level 5/11/17 power spikes. There are perhaps a few individual quirks that could be addressed here and there, but largely there's no harm.


4. Increase ability scores to multi class?
No. In fact, remove them entirely. It's a pointless mechanic.


5. Which classes are mostly dipped for 1 lvl
For one level? Probably Cleric (for domain powers), Fighter (for Fighting Style), and Rogue (for Expertise). If you expand up to level 2, add Paladin (for Smite) and Warlock (for Eldritch Blast/Invocations).

[Keravath]

Most multi classing has an opportunity cost.

5th level sorcerer vs 3rd level sorcerer/2 level warlock

5th level sorcerer has 3rd level spells.
6th level sorcerer has the 6th level archetype feature
7th level sorcerer has 4th level spells (a 5sorcerer/2warlock just gets the 3rd level spells)

Although this depends a lot on class features and in some cases people believe that some classes lack higher level desirable features ... so they sometimes multiclass.

On the other hand, multiclassing is sometimes used for roleplaying reasons. A drow who really wants to be an entertainer/bard but who starts with two levels of warlock to get the invocation allowing the casting of disguise self at will so that they can more effectively travel and interact with the surface world.

I have a few mutliclass characters and there are always times when they feel weaker or less effective than pure classes. ( A level 4 arcane trickster rogue/ 3 fey bladelock vs the level 7 pure warlock in one campaign I am playing ... the pure warlock with PAM typically does a lot more damage than I do).

So multiclassing can add options without necessarily creating "OP" characters.

[AHF]

Outside of the Coffeelock example which seems like the potential for a truly broken mechanic, a one level dip just doesn't seem particularly OP to me. In the absence of it being OP, I would simply require players to have a character/story reason for it. For casters, a short dip delays spell progression and any 1 or 2 level dip deprives the character of an ASI in the long-term and delays the next ASI. For non-synergistic dips (i.e., where main stat for the dip is not one the player would optimize anyway) the cost is even higher.

Not a big problem in the games or builds I've seen (other than Coffeelock which I've not seen in play but seems OP based on the theoretical scenarios in threads on this board).

[Crgaston]

Edit: Wow, stepped away from the computer while typing and in the meantime got multi-ninjaed.

No, it's fine as is. Multiclassing has penalties built in already by way of ASI's being tied to class levels and loss of advancement in the primary class.

Players have different skill levels. Someone who is good at building a character and playing to the strengths of that build is generally going to be more effective than a player who doesn't build well and play the strengths of their build. Changing MC rules won't fix that.

That said, a Warlock2/SorcererX spamming Quickened Agonizing Repelling Eldritch Blasts or a Hexblade2/Paladin2/SorcererX Spamming Quickened BB Smites are easy to make OP in most games once they get into Tier 2 or 3. If there are specific builds causing problems (or that might potentially cause problems) at your table, they can be addressed simply through a Session Zero conversation or through DM adjustments to the adventuring day or types of foes you face. It shouldn't be necessary to penalize players wanting to competently create or play a certain concept by making further restrictions on multiclassing in general.

To address your last question...
Best one-level dips are dependent on the type of character you're building.

Some concepts, as Maxillian points out, rely on multiclassing to even exist, or to function at a competitive level. A non-Hexblade Bladelock needs a level of fighter. An Armored Wizard needs Fighter or Cleric.

Any build that relies on a skill check, such as a grappler or scout, strongly benefits from a level of Rogue. A Bard or Rogue can benefit greatly from the extra cantrips, spells and AC of a 1 level Dragon Sorcerer dip. One level of Monk is great for certain Druid, Cleric or Dex Fighter builds

In all those cases, though, the tradeoff is delayed ASI/feats and delay of higher level features. And those delays are usually going to sting. Your single-classed party members will have maxed primary stats, new higher level spells and abilities for a large number of sessions before the multiclassed character gets them. Especially since, specifically for Fighter or Rogue, if you're dipping 1 you'll probably want to go ahead and take 2 at some point, and then you're even more behind in your primary class.

[Willie the Duck]

For one level? Probably Cleric (for domain powers), Fighter (for Fighting Style), and Rogue (for Expertise). If you expand up to level 2, add Paladin (for Smite) and Warlock (for Eldritch Blast/Invocations).

Also single-level Hexblade Warlock dip by other cha-based classes also seems to be a thing. Overall, it is the cha-based classes which seem to cause most of the bad-taste-in-the-mouth I and most of the people I talk to IRL have for multiclassing.

[Grod_The_Giant]

Also single-level Hexblade Warlock dip by other cha-based classes also seems to be a thing. Overall, it is the cha-based classes which seem to cause most of the bad-taste-in-the-mouth I and most of the people I talk to IRL have for multiclassing.
Forgot Hexblade. Yeah, that's a big one too.

If Cha-based multiclassing is a big issue, switch Warlock to bring Int-based. That leaves a big hole in the Paladin/Sorcerer/Warlock trinity.

[Willie the Duck]

If Cha-based multiclassing is a big issue, switch Warlock to bring Int-based. That leaves a big hole in the Paladin/Sorcerer/Warlock trinity.

Probably does something crazy to wizards, but sure. Although paladin-sorcerer combos still exist. Admittedly, those aren't really overpowered (except in a very '15-minute workday'-favoring game world which undoubtedly has additional problems) so much as it's really annoying that people feel they have to do choose it in the first place.

The overall issue with the Cha-based classes is how different functions of certain class mechanics (the sorcerer spell point system, the paladin spells-into-smites mechanic, the warlock 2-level-ranged-at-will-fix, the hexblade MAD fix work so much better for some class other than the one it is affixed to. We can propose lots of different solutions, but I think there could have just been a little more guidance in the multiclass system saying, in effect, 'this game design did not constrain itself such that multi-classing would not have the possibility of creating potential builds which would or could be disruptive in some way. Even if you the DM decide to include this optional Multi-Classing rule, you should not feel constrained from excluding combinations you find are attempting to abuse this freedom.' I don't know if that would have flown amongst the most fervent optimizer crowd, but it could be sold as a CYA against things like Coffeelock (which I think probably-likely everyone acknowledges as unintended cheeze).

[Sigreid]

I find if someone in my group is pushing the cheese too far the only necessary solution is "don't be a prick". Issue resolved.

[Mortis_Elrod]

i see no reason to discourage optimization, or multi-classing.

If you're having problems its usually because of player behavior, lack of prior communication of intended playstyle, or a misunderstanding on how the game functions. In all of these cases, making multiclassing harder/annoying/useless is not going to solve the issue.

[Corsair14]

I think a return to 2nd edition's multiclassing is what is needed. Any of these 1 level dips is very obviously for optimizing. This game isnt a competition. I think we should go back to where you multi-class at first level. You divide your exp evenly between the three classes as you get it. Powergaming solved. Spontaneous becoming a new class in the middle of a dungeon solved. Wizard's years of research to become level one while a PC can suddenly just become level one with no training at all solved.

[Theodoxus]

I think a return to 2nd edition's multiclassing is what is needed. Any of these 1 level dips is very obviously for optimizing. This game isnt a competition. I think we should go back to where you multi-class at first level. You divide your exp evenly between the three classes as you get it. Powergaming solved. Spontaneous becoming a new class in the middle of a dungeon solved. Wizard's years of research to become level one while a PC can suddenly just become level one with no training at all solved.

Tried that - doesn't work well with the way class and feats have been split out since 2nd Ed.

It's why I jumped at the feat option instead. Works similarly (especially if you take a second class at 1st level) but doesn't lead to weird things like Three classes of 14th level when the rest of the party is single classed 20th level.

I'll just jump in and state that Dual Classing doesn't work in 5E either - asking someone to stop playing with their old toys until their new toys are higher level is just a paragon of frustration - not that it was a great idea in AD&D either... we just didn't know differently.

ETA: I also was very interested in recreating the various XP tracks based on class - such that Rogues leveled faster (required less XP) than Wizards... it exacerbated the mutliclass problem (Having a Fighter/Mage/Thief being 14/12/16 when the rest of the party was 20th), but it did make otherwise less powerful classes (re: non-casters) more powerful - as you were hitting your stride well before the wizards were tossing Fireballs and making you feel useless...

But, when it comes to figuring out how quickly things should develop in relationship to each other... I find I'm not a very smart man. I originally considered using the same ratio as in AD&D... but the game has radically changed, and classes are more balanced, so a rogue quickly became THE choice to play...

I think PF2 could go that route, since they killed the exponential XP needed per level and went with a flat 1000 for each one. A rogue requiring 900, a wizard 1100 and everything in between could work (just as an example, see above about not being a very smart man)...

[Grod_The_Giant]

Tried that - doesn't work well with the way class and feats have been split out since 2nd Ed.
I expect you'd be looking at something more like Pathfinder's Variant Multiclassing (https://www.d20pfsrd.com/classes/character-advancement/#Variant_Multiclassing), where you lose some of the standard progression of your main class in exchange for picking up some key features of the secondary one. In 5e terms, I... guess you'd have to either have "class-in-a-box" feats, which would almost certainly be crazy strong, or generic "cleric/fighter/wizard/..." subclasses that can be spliced in place of a class' normal subclass, which would be all kinds of a headache to balance and fluff.

The alternative, I guess, would be something like 3.5's gestalt, where you get all the features of both classes, taking the better spell/HD/skill choice/etc, paired with some sort of experience penalty. Though again, unexpected synergy (hello sorlock-in-a-box!) makes balance tricky-- a Paladin//Sorcerer is a totally different beast from, oh, a Barbarian//Rogue.

All in all, I think "you wind up X levels behind in your main class for Y levels of your secondary class" is probably the simplest and most elegant.

[Keravath]

The only real example of rules abuse is the "coffeelock". However, if a DM lets a player get away with it then that is a conscious decision on the part of the DM to allow it. In general, I'd expect most DMs just to say No and ask the player to get over it since the sorcerer warlock is already good enough without infinite spell slots.

Other than that ... multiclassing by itself isn't usually a problem.

There are also some combinations of feats that in my opinion can introduce significantly greater player power discrepancies than multiclassing (though even these differences are relatively small compared to previous versions of the game).

[MaxWilson]

1. Is multiclassing for 1 lvl dip or even 2,3 lvls optimizing? Or combo optimizing and backstory?
2. Should multiclassing be revamped to stop the 1-3 lvl dip?
3. If so, how should it be changed? No less than 1-2 levels below others?
4. Increase ability scores to multi class?
5. Which classes are mostly dipped for 1 lvl

If you want to eliminate "dips", you could always write your own multiclassing rules. Here are mine, based on AD&D multiclassing:



Concurrent multiclassing is an option. With concurrent multiclassing, you can advance in two classes or three at the same time, e.g. you could be a 10th level Battlemaster/Necromancer with the abilities of both a 10th level Battlemaster and a 10th level Necromancer. See http://bluishcertainty.blogspot.com/2017/01/5e-old-school-multiclassing-rules_29.html for more details.

[AHF]

I think a return to 2nd edition's multiclassing is what is needed. Any of these 1 level dips is very obviously for optimizing. This game isnt a competition. I think we should go back to where you multi-class at first level. You divide your exp evenly between the three classes as you get it. Powergaming solved. Spontaneous becoming a new class in the middle of a dungeon solved. Wizard's years of research to become level one while a PC can suddenly just become level one with no training at all solved.

I think this would actually be a lot more powerful. Take someone with 130k experience, enough for a character to reach level 13. Using the 2e system, the problem is that you get to take advantage of the leveling scale in multiple classes. So that character gets 13 levels of a single class while the multiclass gets either 10 levels of 2 classes or 8 levels of 3 classes. Ignoring all the interplay 2e had with racial restrictions and class combinations (dwarves and halflings can't be mages, monks can't multiclass, etc.), I'd definitely think long and hard before letting people go down this path.

Half-Elf Paladin Options (3 classes for the fun of it):
(1) Single Class - Level 13 Half-Elf Ancients Paladin
(2) 5e multiclass - Level 7 Ancients Paladin, Level 6 Draconic Sorcerer
(3) 2e multiclass - Level 8 Ancients Paladin + Level 8 Draconic Sorcerer + Level 8 Swashbuckler Rogue (probably have to think about Elvish Accuracy to combine with that Swashbuckler advantage for better odds at critting that smite/sneak attack - just using a familiar does this too)

or

Elvish Eldritch Knight Options (2 classes only):
(1) Single Class - Level 13 Eldritch Knight
(2) 5e multiclass - Level 7 Eldritch Knight + Level 6 War Magic Wizard
(3) 2e multiclass - Level 10 Eldritch Knight + Level 10 War Magic Wizard


Etc.

I think this would seriously increase the power building shenanigans even if you started building in some of the same "rails" that 2e had (like no subclasses for people who multiclass, etc.).

[MaxWilson]

I think this would actually be a lot more powerful. Take someone with 130k experience, enough for a character to reach level 13. Using the 2e system, the problem is that you get to take advantage of the leveling scale in multiple classes. So that character gets 13 levels of a single class while the multiclass gets either 10 levels of 2 classes or 8 levels of 3 classes. Ignoring all the interplay 2e had with racial restrictions and class combinations (dwarves and halflings can't be mages, monks can't multiclass, etc.), I'd definitely think long and hard before letting people go down this path.

Half-Elf Paladin Options (3 classes for the fun of it):
(1) Single Class - Level 13 Half-Elf Ancients Paladin
(2) 5e multiclass - Level 7 Ancients Paladin, Level 6 Draconic Sorcerer
(3) 2e multiclass - Level 8 Ancients Paladin + Level 8 Draconic Sorcerer + Level 8 Swashbuckler Rogue (probably have to think about Elvish Accuracy to combine with that Swashbuckler advantage for better odds at critting that smite/sneak attack - just using a familiar does this too)

Note that option #3 (old-school multiclassing) gives you a lower proficiency bonus, weaker cantrips, and much less HP than option #2 (5E multiclassing, which is really custom class construction). +3 proficiency instead of +5, 2d10 cantrips instead of 3d10, ~60 HP instead of ~96 HP. In return you get more powerful spells and rogue features (4d6 of sneak attack/Uncanny Dodge/Expertise/Cunning Action/Evasion). In most ways I think that is a worthwhile trade and I would take it, but it's not a no-brainer.

Furthermore, the old-school multiclassing isn't just competing with Paladin 7/Dragon Sorc 6. You're also competing with 5E options like Paladin 6/Divine Soul 5/Hexblade 2, which is more SAD (more efficient) and gives you awesome armor class AND Quickened Agonizing Repelling Eldritch Blast for up to 6x d10+d6+5 (84) damage per action, which makes the old-school multiclassing rogue's bonus 4d6 (14) damage look pretty small in comparison.



Elvish Eldritch Knight Options (2 classes only):
(1) Single Class - Level 13 Eldritch Knight
(2) 5e multiclass - Level 7 Eldritch Knight + Level 6 War Magic Wizard
(3) 2e multiclass - Level 10 Eldritch Knight + Level 10 War Magic Wizard

Etc.

I think this would seriously increase the power building shenanigans even if you started building in some of the same "rails" that 2e had (like no subclasses for people who multiclass, etc.).

Again, the relevant comparison is to look at what people actually do with 5E multiclassing, which is less likely IME to be EK 7/War Magic Wizard 6 than Forge Cleric 1/War Magic Wizard 13. It's far from obvious whether that's better or worse than EK/War Mage 10--you're missing out on Extra Attack and the ability to attack for damage + Action Surge a spell at disadvantage to enemy saves from Eldritch Strike, but you're missing out on Simulacrum and Forcecage.

Overall it looks like allowing old-school multiclassing is different, but not more powerful, than 5E multiclassing as long as you don't double up on ASIs and class features. And it results in more organize, archetype-driven characters and less focus on planning "builds". Personally I think it's an improvement.

[ATHATH]

https://1d4chan.org/wiki/Stormwind_fallacy

Side note: If you don't like the Coffeelock cheese, just houserule that you can't burn Pact Magic slots for Sorcery Points. You don't have to restrict/nerf/ban all multiclassing or dipping or something like that.

[AHF]

Furthermore, the old-school multiclassing isn't just competing with Paladin 7/Dragon Sorc 6. You're also competing with 5E options like Paladin 6/Divine Soul 5/Hexblade 2, which is more SAD (more efficient) and gives you awesome armor class AND Quickened Agonizing Repelling Eldritch Blast for up to 6x d10+d6+5 (84) damage per action, which makes the old-school multiclassing rogue's bonus 4d6 (14) damage look pretty small in comparison.

Then complete the comparison with Paladin 8 / Sorcerer 8 / Warlock 8.

[AHF]

Again, the relevant comparison is to look at what people actually do with 5E multiclassing, which is less likely IME to be EK 7/War Magic Wizard 6 than Forge Cleric 1/War Magic Wizard 13. It's far from obvious whether that's better or worse than EK/War Mage 10--you're missing out on Extra Attack and the ability to attack for damage + Action Surge a spell at disadvantage to enemy saves from Eldritch Strike, but you're missing out on Simulacrum and Forcecage.

Overall it looks like allowing old-school multiclassing is different, but not more powerful, than 5E multiclassing as long as you don't double up on ASIs and class features. And it results in more organize, archetype-driven characters and less focus on planning "builds". Personally I think it's an improvement.

I used recommended builds from the multiclassing guide as the base for 5e. I’d say for any reasonably balanced build the old school is better.

A dip build needs to be compared against 2e dual classing for a better comparison.