Multi-class rules ideas

[5eNeedsDarksun]

So, I like the idea of multi-classing and would like to keep it but, as seems to be the case at a lot of tables, I've seen a lot of 1-2 level dips that have mostly served to boost power, and not enhance role playing. So before we get another campaign going I'm trying to come up with some variant rule that would at least encourage some commitment to both classes if that's what people want to do.

So, I was thinking:

... that secondary class(es), if any, must be brought up to 1/3 rounded down of primary class at the earliest opportunity. In short, once primary class hits level 6 all other secondary class(es) need to go to level 2. If primary is 9, others need to go to 3, and so forth. Hopefully that's clear enough. It still lets characters get off the ground by going 5/1 or 6/1 in whatever order they want, which I think is good so that people don't feel gimped early by not getting level 5 soon. But, if they have multiclassed at that point they need to continue to devote some levels to secondary class(es) as they level.

Anyway, I was wondering if the forum had other thoughts on guidelines that might accomplish much the same thing.

[PhoenixPhyre]

Alternatively--junk the whole "generic rules" part except to do things like adjudicate how Extra Attack stacks (or doesn't).

Each class says exactly what you get at each level. Put it in a sidebar. Make each class responsible for what happens when it's a secondary class. This doesn't have to match what the single-classed version gets at those levels--in fact, it shouldn't. It should be delayed or flat out missing chunks. They can be choices (like saying "when you take Cleric N as a secondary class, you get the level 2 Channel Divinity options from your chosen domain"), but no even pretense that you're getting the full class level.

[Rukelnikov]

Alternatively--junk the whole "generic rules" part except to do things like adjudicate how Extra Attack stacks (or doesn't).

Each class says exactly what you get at each level. Put it in a sidebar. Make each class responsible for what happens when it's a secondary class. This doesn't have to match what the single-classed version gets at those levels--in fact, it shouldn't. It should be delayed or flat out missing chunks. They can be choices (like saying "when you take Cleric N as a secondary class, you get the level 2 Channel Divinity options from your chosen domain"), but no even pretense that you're getting the full class level.

What is your secondary class? Just the second one you take or the one in which you have the least levels?

Case A - You spent your elven youth studying Combat, and took a level in fighter, then spent 3 centuries studying magic, are now a Ftr1/Wiz 19, but since Wiz is your "secondary" class, you are a worse wizard.

Case B - You have to be updating your build every time one of your classes surpasses the other, not as bad, but still kinda weird.

[PhoenixPhyre]

What is your secondary class? Just the second one you take or the one in which you have the least levels?

Case A - You spent your elven youth studying Combat, and took a level in fighter, then spent 3 centuries studying magic, are now a Ftr1/Wiz 19, but since Wiz is your "secondary" class, you are a worse wizard.

Case B - You have to be updating your build every time one of your classes surpasses the other, not as bad, but still kinda weird.

Whatever you picked at level 1 is your primary. Anything else is secondary. With this design, you don't have to play funky back-story games or plan out which comes first--pick the one you want more of as your first class. Everything else is secondary. Fighter 1/wizard 19 isn't fundamentally different from wizard 1/fighter 1/wizard 18 in universe. And yes, that makes Case A bad as stated. Which I'm totally ok with. Because your case doesn't even describe a PC at all. The whole "dip one level for that fancy feature" thing (even if by "dip" you mean "start that way but immediately transition out") is a sign that something's fundamentally broken about the way the game is representing stuff.

Personally, I'd go all the way back to the drawing board and just kill level-by-level multiclassing entirely. It's fundamentally ill-suited for a class/level game. Instead, do something like a three-headed approach that all basically boils down to "embrace the homebrew".

1. Make writing new classes fairly simple. So for your deeply involved multiclass builds (mostly equal levels) you and the DM can spin up a properly-tailored base class. Don't even have to make sub-classes because it's fundamentally throw-away.
2. Make writing new subclasses fairly simple. So for your less-equal but still more than a dip builds (6/14, say), you can just spin up a properly-tailored sub class for the "bigger" one.
3. Have each class declare a set of (possibly chained) "feats" you can take to do those "just want that one feature" thing.

[animorte]

I like going back to the drawing board on a level-by-level, state exactly what is acquired. Balancing that out is the problem, but we're already discussing a balancing issue.

This is exactly why I prefer something of a more modular system.

I've thrown up the table several times showing just how beautiful the progression is with subclasses and ASIs covering the entire even levels (except that pesky 19). While all the odd levels cover spell progression or martial progression.

With that thought in mind:

At the first level of multi-class, gain the specified class feature. Each odd level, you should have the opportunity to acquire the appropriate martial feature or maintain your spell-casting progression. You don't get both. For spells, you must progress in order; martial features, must be chosen in order as well.

Either way, subclass features should do one of two things: a) restricted to one subclass or b) gained just as spell progression vs martial features (as stated previously).

[RogueJK]

What the OP is suggesting is broadly similar to what D&D tried back in 3E. In that edition, they had greatly loosened the rules on multiclassing compared to 2E, but instituted a multiclassing XP penalty mechanic to disincentivize dipping into other classes willy-nilly.

Basically, if you multiclassesd, all class levels had to stay within 1 level of each other or you suffered a hefty 20% XP penalty. The only exception was each race was assigned a "favored class", and levels in that class didn't count towards the penalty calculations.

So, for example, you could have a Fighter/Wizard gish character, but you'd normally need to be something like a Fighter 1/Wizard 2 progressing on up to Wizard 6/Fighter 5, etc., or else you'd suffer the consequences of the XP penalty.

(Though if you were of a race who had Fighter or Wizard as their Favored Class, you could have a Fighter 1/Wizard 10 or Wizard 2/Fighter 6 or whatever.)

[elyktsorb]

Hopefully that's clear enough. It still lets characters get off the ground by going 5/1 or 6/1 in whatever order they want, which I think is good so that people don't feel gimped early by not getting level 5 soon. But, if they have multiclassed at that point they need to continue to devote some levels to secondary class(es) as they level.

Anyway, I was wondering if the forum had other thoughts on guidelines that might accomplish much the same thing.

I thought single level dips were just as bad? Why would you make a system that encourages a single level dip, but punishes a 2 level dip? Sure, people will stop dipping 2 levels but not one just for the purposes of extra stuff that doesn't matter thematically.

Also punishing for people who aren't multiclassing in an optimal way in the first place.

[5eNeedsDarksun]

I thought single level dips were just as bad? Why would you make a system that encourages a single level dip, but punishes a 2 level dip? Sure, people will stop dipping 2 levels but not one just for the purposes of extra stuff that doesn't matter thematically.

Also punishing for people who aren't multiclassing in an optimal way in the first place.

Maybe I didn't explain it that well. The idea is that you can't dip 1 level, provided your campaign is going to higher levels (which ours generally do). Once you hit level 6 in your primary class you'd have to take (minimum) a 2nd level in any other classes. At 9 you'd take a 3rd, and so forth, so if you multiclass you're making a commitment to continue to follow both paths.

[Rukelnikov]

Whatever you picked at level 1 is your primary. Anything else is secondary. With this design, you don't have to play funky back-story games or plan out which comes first--pick the one you want more of as your first class.

That's exactly the opposite, you have to go in knowing beforehand what you want to end up being, cause whatever you pick first is what you are gonna be all your career, its terrible from a narrative perspective.

Personally, I'd go all the way back to the drawing board and just kill level-by-level multiclassing entirely. It's fundamentally ill-suited for a class/level game. Instead, do something like a three-headed approach that all basically boils down to "embrace the homebrew".

1. Make writing new classes fairly simple. So for your deeply involved multiclass builds (mostly equal levels) you and the DM can spin up a properly-tailored base class. Don't even have to make sub-classes because it's fundamentally throw-away.
2. Make writing new subclasses fairly simple. So for your less-equal but still more than a dip builds (6/14, say), you can just spin up a properly-tailored sub class for the "bigger" one.
3. Have each class declare a set of (possibly chained) "feats" you can take to do those "just want that one feature" thing.

I think I'd be okay with the feat approach, but anything clunky like "well you passed the treshold of levels allowed in your multiclass, so now you have to review the whole build" stuff is an automatic sign of something not working.

What the OP is suggesting is broadly similar to what D&D tried back in 3E. In that edition, they had greatly loosened the rules on multiclassing compared to 2E, but instituted a multiclassing XP penalty mechanic to disincentivize dipping into other classes willy-nilly.

Basically, if you multiclassesd, all class levels had to stay within 1 level of each other or you suffered a hefty 20% XP penalty. The only exception was each race was assigned a "favored class", and levels in that class didn't count towards the penalty calculations.

So, for example, you could have a Fighter/Wizard gish character, but you'd normally need to be something like a Fighter 1/Wizard 2 progressing on up to Wizard 6/Fighter 5, etc., or else you'd suffer the consequences of the XP penalty.

(Though if you were of a race who had Fighter or Wizard as their Favored Class, you could have a Fighter 1/Wizard 10 or Wizard 2/Fighter 6 or whatever.)

Yeah, I think the idea to disincentivize dips was ok, but the rules made it so harsh it was almost akin to a prohibition (i.e: In all my years playing 3.x I've never seen a character that incurred the XP penalty, cause no one would allow their build to reach that point).

And then they screwed the rule by making PrC's not count, which made everyone just look for ways of classifying for multiple PrC's since not only were they usually more powerful for your build than a base class, but also they didn't incur XP penalties.

Maybe I didn't explain it that well. The idea is that you can't dip 1 level, provided your campaign is going to higher levels (which ours generally do). Once you hit level 6 in your primary class you'd have to take (minimum) a 2nd level in any other classes. At 9 you'd take a 3rd, and so forth, so if you multiclass you're making a commitment to continue to follow both paths.

How would multiclassing work, after 6th level?

Say I'm a Paladin2/Sorcerer6 and I want to take a new class, can I? I'd be breaking the rule of below 1/3rd of the main class.

[PhoenixPhyre]

That's exactly the opposite, you have to go in knowing beforehand what you want to end up being, cause whatever you pick first is what you are gonna be all your career, its terrible from a narrative perspective.

Ok, you can declare which one's primary. But only once, when you take the second class.

[Rukelnikov]

Ok, you can declare which one's primary. But only once, when you take the second class.

I don't think I'll ever understand how someone that seems to favor narrative in RPGs also favors a railroad approach to classes.

[PhoenixPhyre]

I don't think I'll ever understand how someone that seems to favor narrative in RPGs also favors a railroad approach to classes.

Classes are intrinsically railroads. That is their entire point for existing--to constrain the narrative to fixed, specific archetypes. If you want a piece-meal, build-a-bear approach...don't use a class-based system at all. Trying to mix the two just makes a mess.

[Rukelnikov]

Classes are intrinsically railroads. That is their entire point for existing--to constrain the narrative to fixed, specific archetypes. If you want a piece-meal, build-a-bear approach...don't use a class-based system at all. Trying to mix the two just makes a mess.

You may be right, but WoD's marits, disciplines, arcanas, etc. are akin to 5 level classes and work in a point buy system, so I believe there is a way to make it work.

[Witty Username]

Martials get a slight nerf as it requires them to take 5th level in their secondary (extra attack stacking).

Hurts the rogue 1 dip, which I would use to RP a non-rogue archtype criminal.

This looks to be like an RP problem looking for a mechanical solution, rather than an RP one.
Taking some level 1 dips:
Paladin 1/sorcerer X
Hexblade 1/sorcerer X
Cleric 1/Wizard X
Artificer 1/Wizard X

Each of these characters have very different thematics and concepts, demand a bit from the players for what the level 1 represents in their character and concept.

Or drop spellcasting in armor outside of the specific classes armor proficiencies. That kills most of them.

[5eNeedsDarksun]

That's exactly the opposite, you have to go in knowing beforehand what you want to end up being, cause whatever you pick first is what you are gonna be all your career, its terrible from a narrative perspective.



I think I'd be okay with the feat approach, but anything clunky like "well you passed the treshold of levels allowed in your multiclass, so now you have to review the whole build" stuff is an automatic sign of something not working.



Yeah, I think the idea to disincentivize dips was ok, but the rules made it so harsh it was almost akin to a prohibition (i.e: In all my years playing 3.x I've never seen a character that incurred the XP penalty, cause no one would allow their build to reach that point).

And then they screwed the rule by making PrC's not count, which made everyone just look for ways of classifying for multiple PrC's since not only were they usually more powerful for your build than a base class, but also they didn't incur XP penalties.



How would multiclassing work, after 6th level?

Say I'm a Paladin2/Sorcerer6 and I want to take a new class, can I? I'd be breaking the rule of below 1/3rd of the main class.

I was thinking any time your primary class is 6th level or over you'd have to raise the other classes to 1/3 as first priority, so in your example, say you took Fighter 1. You'd have to take Fighter 2 before you took levels in Paly, Sorcerer, or another class.

[5eNeedsDarksun]

Martials get a slight nerf as it requires them to take 5th level in their secondary (extra attack stacking).

Hurts the rogue 1 dip, which I would use to RP a non-rogue archtype criminal.

This looks to be like an RP problem looking for a mechanical solution, rather than an RP one.
Taking some level 1 dips:
Paladin 1/sorcerer X
Hexblade 1/sorcerer X
Cleric 1/Wizard X
Artificer 1/Wizard X

Each of these characters have very different thematics and concepts, demand a bit from the players for what the level 1 represents in their character and concept.

Or drop spellcasting in armor outside of the specific classes armor proficiencies. That kills most of them.

Watching Treantmonk's suggested house rules again, including dropping spellcasting outside of classes that get it, was what got me thinking about this. That option goes further than I wanted to, so I was trying to come up with something less restrictive.
The martial stacking I'm aware of, but we generally finish about the end of tier 3 so it likely wouldn't come up for us. I've always told my players they can have something in lieu if they ever doubled up on that anyway.

[diplomancer]

You could solve most of the problems of dipping with the following rule: "you have the armor proficiencies of the class of which you have more levels. If there's a tie, you can choose".

Alternatively, you could tie spell level to armor proficiency. If you're a Cleric 1/Wizard X, you can cast only 1st level spells in armor (upcasting, as usual, counts as a higher level spell). If you're a Cleric 3/Wizard X, level 2 spells, and so on.

Maybe the one common multi-class this would not "solve" is the Hexblade dip for Paladins; but this is more of a problem with how the Hexblade was designed than a problem with Multiclassing.

[Reach Weapon]

So, [...] I've seen a lot of 1-2 level dips that have mostly served to boost power, and not enhance role playing. [...] at least encourage some commitment to both classes [...]

I believe that house rules that are sort of about one thing, but are then couched in different terms seem to fail and frustrate everyone by turns.

It seems to me that you'd be better off with a simple "builds must get DM approval and are expected to be suitably fleshed out people" rule that at least codifies what (I think) you actually want.

If you're dressing it up because your table wouldn't like that, then trying out different dresses doesn't seem like a real fix.

[Arkhios]

What about this:

In place of an ASI, you can choose a class and gain one or two first level class features from that class. For each ASI after the first one used like this, you can choose new class features from the consequent levels, or choose another class and gain one or two first level class features from that class instead, and so on.

This way you'll keep the feeling of having a primary class and a secondary class.

PS. Some of you may have noticed something similar about this idea; yes, I stole it shamelessly from 4th edition.

[stoutstien]

My (WIP) system has two different multiclassing rules to consider. The first is at creation a hero could just chose to be a X/X (or sometimes even x/x/x) class and they have premade progressions listed. Each class has 1-3 very powerful features that are only accessible via being a single class. This isn't really multiclassing as much as a way have a ton of class options built in.

The second is the story/game driven change. This one is 100% optional and is fully in the hands of the GM. These rules are going to be purely suggestive in nature and are designed to prevent someone from, knowingly or not, circumventing the original rules. The big thing is this option will always have costs. big ones. The chart will mostly just list things that cannot be available to any single hero.

* An example of this would be a deity trying to persuade a hero to commit to an oath to them. In exchange they offer power in the form of an alternative class progression but it would prevent them from taking any further "feats" (name pending) that would require partial/full of that class and they could lose some of those nice class specific features.*

For the most part they are going to be included for completion sake. In hoping to have enough built in flexibility in the class system to prevent need for multiclassing.

[Quietus]

In the context of 5e - just talk to your players. Homebrewing a fix is going to feel hamfisted.

In the context of the new D&D playtest, and as something that could be back- loaded if you wanted, I've been thinking they should have five subclass levels instead of four, and include a subclass ability at level 1. You only get that ability if that's your first character level - you can put your cleric armor proficiencies here, or your hexblade charisma to attack. If you are picking up the class later, you get these abilities at level 3, along with what you would normally get at that level. This won't fully stop builds that grab their dip at first level, but it does mean those builds now wait for level 4 to get their main class's full subclass ability suite.

[Zuras]

Alternatively--junk the whole "generic rules" part except to do things like adjudicate how Extra Attack stacks (or doesn't).

Each class says exactly what you get at each level. Put it in a sidebar. Make each class responsible for what happens when it's a secondary class. This doesn't have to match what the single-classed version gets at those levels--in fact, it shouldn't. It should be delayed or flat out missing chunks. They can be choices (like saying "when you take Cleric N as a secondary class, you get the level 2 Channel Divinity options from your chosen domain"), but no even pretense that you're getting the full class level.

This seems like an unnecessary complication. There’s no need to systematically figure things out beforehand, because there’s no way a lone DM can playtest that for balance anyway. Just outlaw standard Multiclassing but allow custom prestige classes.

When a player wants to multi-class, they work out with the DM:

1) What capabilities they want to gain outside of their current class.

2) What is a fair timeline for gaining those abilities, relative to advancing in their current class as well as compared to other classes.

If you go the PrC route, you can focus on the stuff the player wants to accomplish with the dip, and price it accordingly. Otherwise, you still have the same potential min-maxing issues, along with the feature overlap issues. A war domain cleric Multiclassing into fighter actually gets fewer mechanical benefits than a wizard under the PHB rules, which you can take into account if you do it this way.

In practice, I haven’t seen multiclassing lead to major power creep outside of charisma casters abusing smites and attempts at coffeelocks. Some combinations lead to power spikes at specific levels (Moon Druid 2/Barbarian 1 being the most spectacular) but level out, or can be matched by the right Ancestry and Feat combinations.

All the characters I’ve seen or played with really strong Multiclass progressions have either been in classes with multiple dead levels (rangers and barbarians), charisma casters, and players dipping Fighter 2 so they can action surge two spells at the beginning of combat. If it’s causing a problem, changing how smites and action surges work to be less useful with high level spell slots will fix most of the problems.

[Sigreid]

I think a lot of the issues people perceive with multiclassing could be solved with an old AD&D rule: once you change classes, you can't go back.

[diplomancer]

I think a lot of the issues people perceive with multiclassing could be solved with an old AD&D rule: once you change classes, you can't go back.

I think this would help somewhat with 2 or more level dips, but have very little effect on 1st level dips, and what effect it does have for those dips will be bad; they will just start with the dip, which makes it even more jarring from a story perspective: "I'm a Cleric, but not really, as I actually will be a Bard".

[Xetheral]

Classes are intrinsically railroads. That is their entire point for existing--to constrain the narrative to fixed, specific archetypes. If you want a piece-meal, build-a-bear approach...don't use a class-based system at all. Trying to mix the two just makes a mess.

I disagree that classes are intrinsically railroads. A class-based system can be used to constrain the narrative to fixed, specific archetypes, if that's desired, but that's not the only reason a designer, DM, or player might prefer such a system. For example:

First, level-by-level multiclassing in a class-based system leads to characters acquiring abilities in thematically linked blocks. It's perfectly reasonable for someone who wants the flexibility of a "build-a-bear" approach to prefer built-in thematically linked abilities over an al a carte system where abilities are selected individually.

Second, the character design process in a level-by-level class based system is of a fundamentally different nature than (e.g.) a point-buy system. The former tends to consist of a series of well-defined trade-offs, where the latter is inherently more open-ended. It's entirely reasonable for someone to have a preference for the former process over the latter.

Third, counter-intuitively, a level-by-level class based system can sometimes lead to more mechanical diversity in played characters than a point-buy system. In a point-buy system, for any given aspect of a character there is often: (1) an optimal way to maximize that aspect, and (2) an optimal way to get that aspect up to "good enough" while minimizing point expenditure. Accordingly, at an optimized table it's not uncommon to see every PC use one of the same two approaches to mechanically realizing that aspect of their character. By contrast, the packages of abilities in a level-by-level class system mean that everyone can't just get one of the same two abilities by spending fungible points--the opportunity cost of taking the the necessary levels will vary, and the bonus those best-in-class abilities have over existing abilities in their already-selected classes will also vary. In other words, there can sometimes be a wider variety of "optimal" options in a level-by-level class-based system.

It's entirely possible that you don't personally value any of these features of level-by-level multiclassing, or you see the disadvantages of such a system outweighing any such advantages, or you just don't think 5e's approach actually manages to achieve any of these advantages. :) But I hope I've managed to successfully illustrate why someone might elect to use such a system over a non-class-based system, even if they don't want to constrain the narrative to particular archetypes.

[Sigreid]

I think this would help somewhat with 2 or more level dips, but have very little effect on 1st level dips, and what effect it does have for those dips will be bad; they will just start with the dip, which makes it even more jarring from a story perspective: "I'm a Cleric, but not really, as I actually will be a Bard".

I see nothing jarring about either a bard finding religion and joining a cloister or a cleric focusing more on the promotion side of religion by becoming a bard. Different story if said bard never speaks of his religion again.

[diplomancer]

I see nothing jarring about either a bard finding religion and joining a cloister or a cleric focusing more on the promotion side of religion by becoming a bard. Different story if said bard never speaks of his religion again.

What is jarring is that you want to play a Bard, but have to start with a level of Cleric. You can always justify mechanical choices with in-story causes, but this rule would force all those in-story causes to the first or second level... And both these levels are just too short for that sort of story detail.

[Sigreid]

What is jarring is that you want to play a Bard, but have to start with a level of Cleric. You can always justify mechanical choices with in-story causes, but this rule would force all those in-story causes to the first or second level... And both these levels are just too short for that sort of story detail.

Eh, I don't see an insurmountable problem. The easiest is "I'm a cleric looking to be the next Oral Robert's, religion with a show".

[Rukelnikov]

I was thinking any time your primary class is 6th level or over you'd have to raise the other classes to 1/3 as first priority, so in your example, say you took Fighter 1. You'd have to take Fighter 2 before you took levels in Paly, Sorcerer, or another class.

Sounds reasonable.

Quick question though, whats the biggest issue that's trying to be solved here? Is it just armor proficiencies? Cause I think it may just be better to fix that intstead.

[RedGeomancer]

My rule is:

You must progress to subclass level in all classes that you have before adding a level in a new class.

I don't have so much of a problem with 1 or 2 level dips if it works with the role-playing concept, but want to discourage multiple dips, as well as someone taking the "dip" class first in order to get proficiencies they would lose out on later (e.g. a Wizard X/Fighter 1 starting with Fighter in order to get Heavy Armor Proficiency).

This means that whatever class you start in, you have to earn your subclass (usually at 3rd level) before adding an additional class. If you want the new class to only be a 1 level dip, that's fine, but you can't add a third class if you don't advance your second class to subclass level.

Since some classes reach subclass level at 2nd (e.g. Wizard) and one class reaches it at 1st (Cleric), you could also impose a hard cutoff like you have to reach 3rd level in all classes that you have before adding a level in a new class. If you really want to limit subclassing, you could even require 5th level, which is the major power boost/tier change level.