D&D 5E Multiclassing--Which and Why?

[mlund]

Multi-classing in D&D has always been a train-wreck of an affair. Half the people are trying to create a concept that they can't get due to poor customization options in character creation. The other half are twinks trying to power-game. 3E multi-classing certainly made the jobs of both these core constituencies easier, much to the detriment of everyone else.

Front-loaded classes + 3.X multi-classing is a hot mess. The only thing that balanced it in 3E at all was that dedicated casters were so unbalanced and quadratic that stepping out of them for other classes that drained your Caster Level and Spell Progression was sub-optimal. Meanwhile, no amount of level-dipping for feats and saving throw bonuses would ever put a non-caster on par with a proper caster so it was largely a moot point.

With properly balanced classes front-end loaded multi-classing 3E style is a train-wreck. AD&D multi-classing creates rough power deficits that can't be justified without racial level caps. The alternatives 4E gave us are complicated, though. 4E Feat-based multi-classing is way too dependent on AEDU parallels. Hybrids forced a hard 50/50 split and didn't always support all the supplements or classes properly - plus it's more trouble than they are worth without the Insider character builder.

Appendix-based multi-classing is probably the best way to go that's ever been proposed in D&D. It looks like a lot of work for the design team, though. I'm still curious as to how they'll quantify character-level vs. level of entry vs. number of multi-class levels. It is a lot of variables to balance out.

- Marty Lund

 

[slobster]

I voted 3.x multiclassing, though I agree that it was poorly balanced. But like many other posters, I feel that the freedom it gave the player to really go nuts and explore her options was worth the headaches and work it will take to get it right.

Dipping is one problem. Of all the suggested fixes so far, I like what is apparently the "saga edition" fix the best. When taking level one of a class as a multiclass, you don't get all the front-loaded class features. You choose one. Kind of like hybrid classing in 4E, where you choose a few class features from each of your hybrid classes, instead of getting them all.

Classes that really scale as they level up are another problem for this multiclassing style. Dipping one level of sorceror as a 19th level barbarian gets you a few piddly level 1 spells, hardly worth your time.

Could we balance it so that, instead of taking level one in whatever class you are dipping, you take any level between one and your current character level? So if you were a level 8 paladin who leveled up to character level 9, and wanted to multiclass bard, you could take level 9 of bard and add it on top of your character. That would give you some higher level spells, but probably not the bard's music and stuff, which you would have gotten at lower level. Or you could choose to take bard level 2, where you get inspiring presence (or whatever), willingly trading out higher level benefits for the class features you are actually interested in.

That would allow high level dips of casters and such to be meaningful, but a level 15 fighter/5wizard is still nowhere near as good a caster as a 20 wizard, even though he has some high level tricks.

Of course, bookkeeping as you leveled up three different classes at different class levels vs. character levels would be frakking nightmare.

 

[tuxgeo]

"Lemon curry."

(I didn't vote: I didn't like the choices.)

There have been many good considerations listed in this thread already. It seems clear to me that Wizards have their work cut out for them on this issue.

 

[Li Shenron]

I am not a huge fan of multiclassing, I can definitely live without it.

Generally I am ok with characters having 2 classes or (occasionally) even 3, but beyond that it feels ridiculous to me.

OTOH I prefer unbalanced class combinations, e.g. "fighter with a few levels of wizard", to 50-50 concepts.

And furthermore I think multiclassing is most of the time an idea for later, such as that fighter above later, much later in his career, after being introduced to magic by friends and foes alike, starting to dabble a little bit in magic books, rather than having 2-in-1 characters since the start.

For these reasons, I prefer 3ed style multiclassing, i.e. level-based. Not classic multiclassing because it ends up with evenly balanced classes, and not dual classing because that was so rigid to the point of being silly.

What I didn't like in 3ed style multiclassing was extreme multiclassing (more than 2-3 is extreme enough for me), level-dipping done just for efficiency reasons, and the fact that it worked too well for fighter-types and horribly for casters.

4ed style multiclassing is not bad, but I would not actually call it multiclassing.

 

[Lanefan]

Voted for "classic" but it's still not perfect:
- I'd like to get away from 3e-style additive levels and have each class be independent
- I'd like to see variable XP assignment allowed to each class in a pre-set ratio, so one can be a 75% Fighter-25% Thief if desired
- I'd like to see severe restrictions on how many classes one can be overall during one's career (3 max., 2 is better)
- Some classes, for flavour reasons, should not be allowed to multiclass at all - Monk, Bard, Paladin: I'm looking at you
- And most importantly, if one isn't multiclass right from the start there should be some in-game headaches involved with picking up a new class: half a year of training, that sort of thing

Another thing that's needed is a way to drop a class completely. An example of when this might come up would be an Assassin who has had a change of heart (or alignment) and wants to erase the class. I dreamed up a high level Cleric spell for this; it can only be cast on a willing un-charmed subject, and it erases the class completely along with all assiciated abilities, hit points, etc.

Lanefan

 

[Lord Mhoram]

I prefer 3rd normal, followed very closely by gestalt (which is very close to 1st ed multiclassing).

I like the idea that each level can be a discreet set of ablities, and the character is defined by the abilities they can do rather than a nebulous idea of "Fighter" or "Wizard". It really opened up the kinds of characters you could play.

Yeah there was abuse, but I much preferred that system.

 

[KidSnide]

Flawed multiclassing rules have been part of D&D for nearly as long as armor class.

That having been said, I think there is something to be said for classic AD&D style multi-classing. Conceptually, AD&D multiclass characters are wildly unbalanced. But in practice, I always found that the single class spell casters didn't feel overshadowed by the multiclass spell casters because they were a level or two ahead and had notably superior spell casting powers. Likewise, the single class fighters weren't overshadowed by the multiclass fighters due to weapon specialization and significantly better hit points.

Likewise, I find that multiclass characters with classic rules really do deliver on what they promise. Characters really are full members of both classes. A lot of 3.x characters and 4e hybrid characters can't say that.

-KS

 

[Gadget]

With all the heavy emphasis on the narrative and story role of classes the designers have made lately, I was disappointed to learn they were going toward a more 3e-centric approach. At some point, why even have a class based system if you're just going to reduce each level to a set of pre-packaged abilities you can choose at level up? I admit that on paper it sounds elegant and efficient, but in practice I think it ended up being impractical and cumbersome: your character was either ineffective or you had cherry-picked the right stuff to make your killer 'build' combo. I guess the info is too vague to pass judgement on yet, but I hope they make it work well, and not just trying to patch up 'the spell caster mulit-class problem' either.

 

[Tony Vargas]

3e's multi-classing was a quantum leap in character customizeability and a great innovation. As with healing surges & hit dice, though, it looks like 5e is going to try to improve on something good, by averaging it out with something not quite as good. 3e class/levels were like modular building blocks, they quite efficiently and elegantly allowed you to build to concept using bits of each class. What's described here also builds-in some of the 'averaged out' half-classes of classic D&D (you multiclass Fighter, you get half the hps of a fighter of the same level, you don't get the THAC0 of a character of your total exp) and 4e Hybrids. Those worked, in a way, thanks to the complexities of level advancement, but they lacked elegance, consistency, and flexibility.

3e multiclassing was a great idea, but it had a couple of little blind spots in the implementation. Like caster level. BAB increased for every class, even the wizard. Caster level didn't. If caster level had advanced like BAB, with non-casters getting 1/2 progression, hybrid casters like paladins 3/4, and casters full, multi-classing to and among caster classes would have been much more practical. Fixing those would not have been hard. I could have been done with 3.5, but it wasn't. 5e has changed some basic advancement around and has a sort of fetish for disparate sub-systems for each class, so it won't be that simple to just take the 3e aproach and drop it in. Without quasi-uinversal progressions like BAB and Caster Level or an actually universal progression like pre-E's (that is, with bounded accuracy), there's no way to let different classes build on eachother. Instead, MC'ing will give you a collection of unrelated abilities. You'll have a spell from list A and a spell from list B and a CS die and an SA die, and you'll be like a party of 1st level characters sharing a single set of actions.

A better aproach, IMHO, would be to use a common advancement structure, and let multi-class characters pull options from each of their classes at each level. No need to worry about BAB or Caster Level (or the 5e equivalent damage progressions), or make multi-class characters wait multiple levels for the features they want most. Think of 4e Themes and Racial substitution powers. Being multi-classed doesn't need to give you more stuff, just more options. You could potentially retain most of the class balance that 4e achieved and most of the flexibility of 3e modular multi-classing with such an approach. That would be something.

 

[Dannyalcatraz]

As I stated elsewhere, 3Ed style is far and away my favorite form of multiclassing, for a variety of reasons. HOWEVER, I think that 4Ed Hybridizing represents another kind of multiclassing, one that goes back to AD&D, and deserves to be supported.

So I would prefer a system that had both 3Ed free form and 4Ed Hybrids as options.

 

[trancejeremy]

Classic for me.

3.x multiclassing to me is broken on several different levels.

Mechnically, which is why there had to be kludge prestige classes like the Mystic Theurge

Logically. I just leveled up after I was in an adventure as a fighter, so now I can take a level in any other class, despite no training or practice.

And gaming wise, it just encourages character optimization. People don't play characters, they play "builds".

 

[I'm A Banana]

Am I the only one with a soft spot for Gestalt?

"Here. Have another class. Think it's too powerful? Give it to your monsters, too. Throw bigger things at them. Too much for you? Can't handle the awesome? Well, go have a lie down, Nancy, these magical gumdrop elves are for CERTIFIED BADASSES ONLY."

Okay, yeah, not the easiest thing to balance in the world, but a lot of fun anyway.

 

[Dannyalcatraz]

Hybrids are kind of like Gestalts on Valium, and I like them that way.

 

[Obryn]

For me, D&D is about archetypes. Class should be the #1 defining feature of a D&D character. 3.x is the outlier here; while it's a fun pseudo-point-buy system, I prefer it when multiclassing is minimal - if it's even present at all. This was even the case in my 3.x days.

Ideally, I'd like to see only feat-based multiclassing (like, spend a feat to get a class feature with restrictions), or pre-packaged multiclasses built in a single class progression (like a fighter-wizard "bladesinger", a rogue-fighter scoundrel, and so on).

-O

 

[FireLance]

I've suggested before that there might be some scope to combine the best of 3e-style multiclassing and 4e-style feat-based multiclassing. Before you can start gaining levels in another class, you need to take one or more feats which will give you some of the new class's abilities. Taking a multiclass level then becomes something like taking a prestige class - you need to meet certain prerequisites first. So, before a fighter can take a wizard level, maybe he needs to take the Arcane Dabbler feat. Before a wizard can take a fighter level, maybe he needs to become proficient in at least one martial weapon.

 

[Magil]

I'm surprised there's as many votes for 4E feat-based multiclassing as there are, as generally the reception to it I see on the forums is rather cold. Still, I think it's my favorite brand of multiclassing, cutting out the gamey potential for abuse of 3rd edition multiclassing/gestalt without the ease of falling into a trap that is the 4th edition hybrid system. I'm too unfamiliar with the "classic" options to comment on them. I feel like it's a decent compromise to keeping character ideas focused, while still allowing a reasonable amount of customization.

I don't think such a system would work well in DnD Next's current iteration, however, due to feats coming every 3rd level instead of every 2nd. Incidentally, I'd like that to change, but I see it as unlikely.

 

[Tony Vargas]

I'm surprised there's as many votes for 4E feat-based multiclassing as there are, as generally the reception to it I see on the forums is rather cold.

It's got the core of a sound enough system - swapping out powers - but it seriously overvalued power-swapping.

I don't think such a system would work well in DnD Next's current iteration, however, due to feats coming every 3rd level instead of every 2nd.

And also because there are no equivalent powers to swap. 4e feat-based multi-classing couldn't even work for all Essentials classes, for instance, it was predicated on that common character progression structure.

 

[slobo777]

I've suggested before that there might be some scope to combine the best of 3e-style multiclassing and 4e-style feat-based multiclassing. Before you can start gaining levels in another class, you need to take one or more feats which will give you some of the new class's abilities. Taking a multiclass level then becomes something like taking a prestige class - you need to meet certain prerequisites first. So, before a fighter can take a wizard level, maybe he needs to take the Arcane Dabbler feat. Before a wizard can take a fighter level, maybe he needs to become proficient in at least one martial weapon.

I like that. Pre-requisite multi-classing at least prevents the weird immediacy of new abilities that come with 3E level dipping.

The D&D Next multi-class systems looks like each class will get deconstructed into a special multi-class version. Which is pretty similar, really, to prestige classes. Add some simple pre-requisites (even just old-school must have 13 Int to multi-class Wizard), and in practical terms, taking a D&D Next multi-class option would be much the same as taking a 3E prestige class.

Add a bonus feat back for Humans, so it can be spent obtaining pre-requisites early, and they become best race for multi-classing.

 

[Wepwawet]

4E style multiclassing feats.

It's far from perfect and too weak, but i like the dipping modularity that it gives PCs.
3.5 multiclassing would be ideal, but that has too many problems. I think that 4E multiclassing feats are a very similar system, but much more limited and controlled. And you have the choice of how much and when you dip.
(But the limit of one multiclass would have to go)

 

[Wepwawet]

BTW, what is classic multi- and dual-classing?