Multiclassing seems to suck

[zazq]

So i see all these builds floating about full of multiclassing and i can't make the numbers add up. Give me a hand.

Assumptions:
There is exactly 10 rounds of combat every single day.
PCs hit the average monster on a dice roll of 11. (+7 for a level 8, +8 for a level 11)
PCs hit for fairly even damage regardless of build. (26 damage per turn at level 8, 41 damage per turn at level 11)
Multiclassing delays both Base Attack progression and Ability Score Increases, resulting in a lower attack bonus at some levels.

Thus:
Multiclassing in such a way that reduces your attack bonus by -1 means you hit on 12s instead of 11s, a 10% decrease in damage, meaning 2.6 at level 8 and 4.1 at level 11.
For the multiclassing to be worth it, it has to provide more damage or damage equivalent effect to make up for the reduced damage.

At level 8, a 2 level dip would have to produce 26 damage over 10 turns, and none of them do. Everybody loves dipping paladin for smite, but 2 smites a day at 2d8 gets you only 18 damage. Granted the damage is front loaded which is good, but i'm not sure its THAT good. You'd need extra spell slots to burn to get to 26, but usually those spell slots are in some way part of your expected 26 damage per turn, so using them up has its own impact on damage output.

So what's the deal?

[Klorox]

How does multiclassing reduce your base attack progression?

[zazq]

erm, its called 'proficiency bonus' in 5e i guess. Your character's proficiency bonus is determined by your highest class level, not your total class level. A level 5 barbarian has a +3 proficiency bonus but a barbarian(4)/fighter(1) has only a +2.

[Tbonefahy]

erm, its called 'proficiency bonus' in 5e i guess. Your character's proficiency bonus is determined by your highest class level, not your total class level. A level 5 barbarian has a +3 proficiency bonus but a barbarian(4)/fighter(1) has only a +2.

That's not how that works, Proficiency Bonus is based on total character level.

[danksteel]

erm, its called 'proficiency bonus' in 5e i guess. Your character's proficiency bonus is determined by your highest class level, not your total class level. A level 5 barbarian has a +3 proficiency bonus but a barbarian(4)/fighter(1) has only a +2.

That's not true at all.


P r o f i c i e n c y B o n u s
Your proficiency bonus is always based on your total
character level, as show n in the Character Advancement
table in chapter 1, not your level in a particular class.
For example, if you are a fighter 3/rogue 2, you have the
proficiency bonus of a 5th-level character, which is +3.� - p163, phb

[Klorox]

Many classes have frontloaded abilities, so you'll see a lot of optimal builds with 2 levels of fighter (fighting style, best armor, action surge), 2 levels of paladin (armor, smiting), 2 levels of warlock (invocations), 1 level of rogue (expertise is 2 skills), ect..

Many classes best abilities come at a slightly later level. Take the paladin, for example; here you have a charisma based warrior who really doesn't get much better as you go past level 6.

They are balanced with the number of spell slots they get, but what happens when, after level 6, you decide to go into sorcerer or bard, a full spellcaster who is also charisma based?

You get a whole lot more, in both quantity and quality, of smites per day.

Multiclassing is an extremely strong option in this game.

[pcamp88]

As was stated above, multiclassing can be very powerful in the hands of a good planner. Some builds can feel like they take forever to come online though depending on your starting level. Not getting to at least level 5 (where every class gets it's first major power bump) in your starting class can leave you feeling weaker than the rest of the party in a lot of situations, but even then it's a case by case basis depending on the build you're going for.

[zazq]

oh, well good job, you answered the question. if multiclassing doesn't delay proficiency bonuses that cuts in half the downsides. It only delays Ability Score Improvements, and only if you don't take 4 levels.

So lets talk about missing those. I'm assuming that everybody uses their first 2 ability score improvements to up their attack paired stat to 20 and missing this would cause the problems already mentioned. After that it is much less of a big deal.

So no multiclassing other than 4/4 until after level 8?

[Kane0]

I dunno man, Fighter/Rogue seems to be as strong if not stronger than it's ever been in prior editions. Caster mixes too.

Proficiency is based off character level
Level 1-3 is often where the best abilities lie
A lot of classes have obvious breakpoints (common ones being 1, 2, 3, 5 and 11 for defining class abilities and second attack but also 4, 6, 7, 8 and 12 due to ASIs and more specific class abilities)
Many classes offer a lot less in terms of raw power at mid to late levels
Many classes get a great deal from pinching abilities otherwise unavailable to them (full caster slots for paladins, short rest slots for sorcerers, reckless attack for rogues, etc)
Spellcasting slots get combined with casting classes (warlock excluded, which offers its own benefits), so the golden rule of 3.X caster multiclassing no longer applies (thou shalt not relinquish casting progression)
Giving up an ASI to get some benefit you would otherwise get from multiclassing (magic initiate, weapon master, armor proficiency, etc) often is not worth all the things that 1 or more levels in another class would do for you instead

[Honest Tiefling]

So no multiclassing other than 4/4 until after level 8?

Briefly did the whole cleric 1/wizard X thing at level 8. As long as your INT is alright and you have a +INT race, it's perfectly fine because you get two expertises, guidance spam, medium armor in exchange for delaying wizard levels by 1. Since spellcasting levels stack, it's really not that bad.

Also, for roles who don't do damage (support and controller), the lack of damage isn't a good argument.

[ZorroGames]

Look here.

http://www.giantitp.com/forums/showthread.php?502248-Ultimate-Optimizer-s-Multiclassing-Guide&highlight=Multiclass+guide

Very helpful for me.

[bid]

I'm assuming that everybody uses their first 2 ability score improvements to up their attack paired stat to 20 and missing this would cause the problems already mentioned.
This is rarely true. Most build will take a feat at level 4 and reach 18 in their primary stat at level 8.

Considering how many dips offer more than a feat's worth, delaying your main progression is often worth it.
- cleric 1 for heavy armor (some domain)
- druid 2 for wildshape
- fighter 1 for heavy armor
- paladin 2 for smiting caster
- rogue 1 for wrestling with athletics expertise
- warlock 2 for agonizing EB
- many 5 for extra attack, perfect for ranged rogue

The elephant in the room is the delayed spell progression, but that won't stop gishes and other martials.

[Fflewddur Fflam]

Completely uninformed and factually incorrect opinions seem to suck.

[Vaz]

Briefly did the whole cleric 1/wizard X thing at level 8. As long as your INT is alright and you have a +INT race, it's perfectly fine because you get two expertises, guidance spam, medium armor in exchange for delaying wizard levels by 1. Since spellcasting levels stack, it's really not that bad.

Also, for roles who don't do damage (support and controller), the lack of damage isn't a good argument.

They only stack for spell slots, not preparation.

But yes, multiclassing does suck. A 10/10 build is ridiculously poor and very slow to come online and introduces arbitrady requirements not there for single classed characters (Dex Paladin, a no-spell Attack/DC melee Spellcaster being completely legal options until you throw in another mulitclass requiring 2-3 13's in scores you might not want)

Any build taking any more than X 3/Y 17 had better know what's really doing, and multiclassing for anything more than a dip seriously hampers any sort of chabuild. Given that WotC have ensured that end game characters aren't really a thing and balance really around 12-13, many classes suffer from not being dip worthy.

It's a dedicated design option that leaves you waiting for 4-5 sessions after a multiclass until your character concept comes online.

5e is D&D, an idiots introduction, built for simplicity and ease, and it was designed intentionally to make it hard for people to optimize outside of a limited range (removing feats, capping ability scores, multiclassing being seperate from how your good your character is and actively makes the meta concept of class an active part of the game, rather than having character concept being representated by character classes).

[CaptainSarathai]

So i see all these builds floating about full of multiclassing and i can't make the numbers add up. Give me a hand.

Assumptions:
There is exactly 10 rounds of combat every single day.
PCs hit the average monster on a dice roll of 11. (+7 for a level 8, +8 for a level 11)
PCs hit for fairly even damage regardless of build. (26 damage per turn at level 8, 41 damage per turn at level 11)
Multiclassing delays both Base Attack progression and Ability Score Increases, resulting in a lower attack bonus at some levels.

Thus:
Multiclassing in such a way that reduces your attack bonus by -1 means you hit on 12s instead of 11s, a 10% decrease in damage, meaning 2.6 at level 8 and 4.1 at level 11.
For the multiclassing to be worth it, it has to provide more damage or damage equivalent effect to make up for the reduced damage.

At level 8, a 2 level dip would have to produce 26 damage over 10 turns, and none of them do. Everybody loves dipping paladin for smite, but 2 smites a day at 2d8 gets you only 18 damage. Granted the damage is front loaded which is good, but i'm not sure its THAT good. You'd need extra spell slots to burn to get to 26, but usually those spell slots are in some way part of your expected 26 damage per turn, so using them up has its own impact on damage output.

So what's the deal?

The deal is, you have made several, blatantly wrong assumptions, your math is haywire and you don't quite understand the rules.

So let's begin:

10 rounds/day.
The DMG assumes 3 rounds per combat, based on their math for damage output. Most parties who bother tracking this kind of thing report back a rough average of about 5 rounds. Per combat. There are about 6-8 Encounters per day. Granted, Encounters might not always be combat. But let's just ignore that. So you're really looking at more like 18-24 rounds of combat per day by the DMG, or 30-40 based on player experience. But ah well, that actually supports your claim.

PCs hit average monster an 11 (11+7)
That's Full Plate AC, friend. Most monsters have a lower AC, as part of a concept called 'Bounded Accuracy.' As monsters get bigger in 5e, their AC stays roughly the same, but their HP grows. You can find relatively low level monsters with very high AC values. You can find boss monsters with lower AC. By the DMG, you don't start hitting AC18 monsters until 15-16th level. Most ACs hover closer to about 15/16, especially at the lower levels where most of the game takes place. So at that point, players with a +3 Ability and Proficiency is pulling about +6ish to hit. So, hitting on 9s or 10s, not 11+.
But that doesn't matter because a +1 is only ever worth +5% on your chance to hit because it's taken out of your chances over 20. Going from an 11 (9/29) to a 10 (10/20) is only an increase from 45% to 50%.

Average Damage by Build is 26
This would be considered DPR, a class designed around doing equal, consistent damage all day long.
The notion that you'd do 26 damage is off. You're looking at something more like ~3.5-4.5 per attack (based on Cantrip scaling).
That is straight sustainable damage. That's the Wizard using his Cantrips, the Fighter swinging his d6-d8 sword. Yes, the Fighter gets his +Stat to damage, but the Casters usually do not. Thus, average. These are the numbers the developers have balanced the game around. Optimizers can usually get this number a bit closer to 5*Attack, but hey - they might be Multiclassing.

So, let's look at math
So, with something like an 11th level character, getting a generous d8 per attack, and hitting (again generously) on a 9+, you're going to see a themselves scoring out:
.55*4.5 = 2.47 damage per attack.
Increasing that to hitting on an 8+ (unlikely, because you've gained 4 levels by that point and CRs have increased accordingly - again, bounded accuracy) you look at:
.6*4.5*3 = 2.7 damage per attack.

Some Rules Issues
Firstly, Proficiency Bonus increases with Character Level, not class.

Secondly, you don't understand the rules for Smiting: you can wait until after you've rolled to hit. Most Paladins only burn the slot on Criticals and most folk looking at damage output will burn their highest level slot. So a whole lot more than "2d8" at 11th level. At least 4d8.

Some Assumption Issues
You're assuming that all builds are designed solely around dealing out consistent damage. They are not. In fact, Fighter and Barbarians are the only ones who really have an even, consistent damage output. If you are only looking at consistent numbers, of course Fighters are king - they get more Attacks per action. Warlocks aren't bad off either, getting Agonizing Eldritch Blasts for a d10+Cha per Fighter Attack as a Cantrip.

You're assumption that fights only last about 1-2 rounds probably comes directly from this, in fact. Paladins running in and blowing their Smites right there on that first attack, Wizards burning off high-level spells like candy, Fighters hammering through their Maneuvers and so on. What you're probably seeing are what we call "5 minute adventuring days," where the party burns all their heavy stuff, and then decides to take a Long Rest and the DM just let's them.

You're also assuming that a Feat or ASI (+1 To Hit, +1 Damage) is somehow worth what you might get from a front-loaded class feature. Barbarian 2 only slows down Feat gain, but gives you at will Advantage as a sustainable option. The combines with Rogue for at will Sneak Attack damage.

Conclusion
The fact is, most classes are not built around dealing sustained damage. They're built to deal bursts of damage. The duration and frequency of these bursts tends to vary, but they're normally built in burst damage.
A party where every member is only throwing their standard Attack Actions and Cantrips for damage is a party on the ropes.

And that is what Multiclass players are doing. They are willing to shift their class somewhere along the Sustain/Burst spectrum. A caster taking levels of Warlock for Short-Rest Recharged and access to Agonizing Eldritch Blast is looking to increase their Sustainable damage at the cost of their Burst damage. A Fighter who dips into Rogue is likely offsetting his Sustainability for Burst capacity.
Also, some players are just trying to tweak a class to their playstyle. Folks who are Multiclassing around Warlocks are probably trying to make the class better in melee, to fit their notion of what a BladeLock should be.

And in some instances, it's very hard to argue with the strength of a Multiclass. Consider for example:
Warlock 2, Sorcerer 6.
They give up spell slots, but tell me what spell deals 8d10+8*Cha in Force damage, and then pushes the enemy 40'.
Because a Sorcerer who burns his recharging Warlock slots for Sorcery Points, and then Quickens his Eldritch Blast attack is getting just that.

That's just one of many Multiclass that can push damage well above what you'd expect for a given level.

Hope that answers "what's up."

[Findulidas]

What I like with 5e is that you can actually stay with any class all the way up to 20 and not feel gimped. You can also multiclass and its not too bad. So you can really do what you want.

[Foxhound438]

So lets talk about missing those. I'm assuming that everybody uses their first 2 ability score improvements to up their attack paired stat to 20 and missing this would cause the problems already mentioned. After that it is much less of a big deal.

So no multiclassing other than 4/4 until after level 8?

personally, no and no.

For the first, you can generally get away with a 16 or 18 for a long time, even in a primary attack stat. This is a lot easier to manage if you're in a group where "magic item shops" exist and you can just spend 500 gp to get a +1 weapon to offset it. You could argue that 20 with the +1 is better, but I could argue diminishing returns. More often than not I see people looking to sacrifice early stats for early feats, hence why Vuman is so popular. It's also easier to manage if you're picking up things in your overall build that mitigate accuracy loss (ie, a paladin using bless and picking up polearm master at 4 or a fighter using archery style picking up crossbow expert at 4).

As far as level splits go, there's basically 2 correct ways to go about it in my opinion: either start with 1 of something and then immediately get 5 of another, or get at least 5 of your main class before any of a second. Reason being that level 5 has a huge power spike for every class except rogue, so you want to get that sooner rather than later. Delaying an ability score increase isn't really that big of a deal in this case, either, because you're still getting relevant features at every level... For example, a monk getting level 8 can get a minor damage boost off of +2 dex, but a monk 5/ ranger 2 picking up a third ranger level gets a pretty close to equal damage boost out of colossus slayer, and has the major damage boosts of hunter's mark and dueling style that a pure monk would never get.

[Lombra]

One level dip of rogue is good for every DEX based character, two levels of warlock are good for any CHA based character, one level of cleric is popular, along fighter, which is worth both in one, two or three level dips.

Yeah one asi can be dalayed/removed depending on which level the character will reach.

[rigolgm]

This is an odd thread. Multiclassing works nicely. The game designers did an amazing job of keeping multiclassing broadly very balanced and also interesting.

If anything, my only objection so far has been a Fighter/Rogue archery multiclass that seemed a bit overpowered (and also boring). Some of the threads about Walrock multiclassing etc also seem either cheesy or like they don't follow 'rules as intended', but I've yet to properly look into them.

I love how my gnome Wizard took his first level in Cleric. Sure, he's one level slower in getting certain upgrades (stat increases, spell level increase etc) and he'll never get the level 20 Wizard ability. But he has good armour/weapon proficiencies, three additional cantrips, more total 'memorized' spells, access to good 1st-level healing spells (which can be cast with higher spell slots, it seems) and he still gets full increases in spell slots, skill proficiency etc. As a Knowledge Cleric, he even started with six skills including double-proficiency in Arcana and Nature, which can make the other Wizards jealous :) And I believe he can use magic items made only for Clerics.

Of course, I still restrict myself by trying to roleplay him so that he's true to his religious faith (in Mystra and Chauntea). Overall, a superb toolbox character.

[Findulidas]

This is an odd thread. Multiclassing works nicely. The game designers did an amazing job of keeping it broadly very balanced and also interesting.

If anything, my only objection so far has been a Fighter/Rogue archery multiclass that seemed a bit overpowered.

I love how my gnome Wizard instead took his first level in Cleric. Sure, I'm one level slower in getting certain upgrades (stat increases, spell level increase etc) and I'll never get the level 20 Wizard ability. But I have good armour/weapon proficiencies, three additional cantrips, more total 'memorized' spells, healing spells and I still get the same quantity of spell slots, proficiency etc. As a Knowledge Cleric, I even started with six skills including double-proficiency in Arcana and Nature, which will make the other Wizards jealous :) And I believe I can use magic items made only for Clerics. Of course, I still restrict myself by trying to roleplay him so that he's true to his religious faith (in Mystra and Chauntea).

He's a superb 'toolbox' character and has a nice rivalry with the much more focussed Sorcerer in our party. I don't want to annoy him too much, though, as Sorcerers can get 'Subtle' metamagic that bypasses my Counterspell! ;)

It still annoys me that other classes easily get as high arcana and even higher than wizards. If someone picks wizard if Im dm im actually considering giving them arcana as expertise at some level.

[rigolgm]

It still annoys me that other classes easily get as high arcana and even higher than wizards. If someone picks wizard if Im dm im actually considering giving them arcana as expertise at some level.

Yeah. Same with Druids potentially having much worse Nature skill than Rogues etc (especially as INT is basically a dump-stat for Druids). The recent Unearthed Arcana update gives a solution - the ability to use feats to give double proficiency (and a bit more): http://media.wizards.com/2017/dnd/downloads/UA-SkillFeats.pdf As with a lot of Unearthed Arcana stuff they seem a bit abusable i.e. giving Fighter or Barbarian grapplers double proficiency in Athletics etc.

[Findulidas]

Yeah. Same with Druids potentially having much worse Nature skill than Rogues etc. The recent Unearthed Arcana update gives a solution - the ability to use feats to give double proficiency (and a bit more): http://media.wizards.com/2017/dnd/downloads/UA-SkillFeats.pdf As with a lot of Unearthed Arcana stuff they seem a bit abusable i.e. giving Fighter or Barbarian grapplers double proficiency in Athletics etc.

Yes, if a druid in my group asked for expertise in nature he might get it as well.

We are using those feats while we are playing right now. The dm houseruled that the bard could have both expertise and that feat tripling his deception proficiency (normally you can only double proficiency). Its a bit broken but at the same time hilarious. Its also his first game ever playing dnd so its not that bad like you would think.

She did ok me picking the feat for athletics in my future fighter lizardfolk pirate grappler build though, will be interesting to see how grappling two people and then biting the heads off will work. Or grappling them and jumping into the lake to drown them.

[Kobard]

It still annoys me that other classes easily get as high arcana and even higher than wizards. If someone picks wizard if Im dm im actually considering giving them arcana as expertise at some level.The skill system (and expertise) is definitely not 5E's strong point.

[bid]

So, let's look at math
So, with something like an 11th level character, getting a generous d8 per attack, and hitting (again generously) on a 9+, you're going to see a themselves scoring out:
.55*4.5 = 2.47 damage per attack.
Increasing that to hitting on an 8+ (unlikely, because you've gained 4 levels by that point and CRs have increased accordingly - again, bounded accuracy) you look at:
.6*4.5*3 = 2.7 damage per attack.
Expected to-hit is .65 or so.
- cantrip (wizard at 11) = 1d8 * .65 * 3 = 8.8
- sword (fighter at 11) = 1d8+5 * .65 = 18.5

[Khrysaes]

Expected to-hit is .65 or so.
- cantrip (wizard at 11) = 1d8 * .65 * 3 = 8.8
- sword (fighter at 11) = 1d8+5 * .65 = 18.5

Warlock is a bit better with agonizing blast.

Sorcerer has it in between warlock and wizard with Elemental Affinity.

Not counting UA, which can grant warlock elemental affinity in Fire and Radiant, and sorcerer another way in lightning.

Your fighter math is missing the *3 for the 3 attacks.

Also consider that Strength is the easiest stat to raise past 20, with multiple ways to do so. the rest require Tomes of ____ I think?

[Spiritchaser]

Yeah. Same with Druids potentially having much worse Nature skill than Rogues etc (especially as INT is basically a dump-stat for Druids). The recent Unearthed Arcana update gives a solution - the ability to use feats to give double proficiency (and a bit more): http://media.wizards.com/2017/dnd/downloads/UA-SkillFeats.pdf As with a lot of Unearthed Arcana stuff they seem a bit abusable i.e. giving Fighter or Barbarian grapplers double proficiency in Athletics etc.

I'd actually suggest that giving fighters and barbarians the ability to more effectively grapple without MC was a needed change.

Rough on casters?
Yes.
Good.

[Maxilian]

Yeah. Same with Druids potentially having much worse Nature skill than Rogues etc (especially as INT is basically a dump-stat for Druids). The recent Unearthed Arcana update gives a solution - the ability to use feats to give double proficiency (and a bit more): http://media.wizards.com/2017/dnd/downloads/UA-SkillFeats.pdf As with a lot of Unearthed Arcana stuff they seem a bit abusable i.e. giving Fighter or Barbarian grapplers double proficiency in Athletics etc.

I actually have no problem with it, i prefer this over having the Fighter or Barbarian MC 2 lvls into Rogue or Bard for the expertise.

[Specter]

Multiclassing only sucks if you don't know what you want. Otherwise, it's as good or better than singleclassing.

The most common denominator of what's good in terms of optimization is damage, so for instance, a Sorcerer/Warlock 2 will always be better than a pure Sorcerer. But you can plan around any pillar of the game and get a good result. If you're a ranger and want more healing, for example, there's little reason not to take Life Cleric.

[Tetrasodium]

Multiclassing only sucks if you don't know what you want. Otherwise, it's as good or better than singleclassing.

The most common denominator of what's good in terms of optimization is damage, so for instance, a Sorcerer/Warlock 2 will always be better than a pure Sorcerer. But you can plan around any pillar of the game and get a good result. If you're a ranger and want more healing, for example, there's little reason not to take Life Cleric.

Life cleric s great for any caster who wants healing or heavy armor & you get cure wounds/bless as free prep domain spells to boot on top of some potentually nifty cantrips

[The-0-Endless]

Looking at average damage output per level exclusively, you speak the truth. However, for almost anything else, whether it be RP, creative use of abilities, really anything other than munchkining is better when used with multiclassing.