D&D 5E (2024) Class and Subclass Design: What Works

[Minigiant]

You're not going to find a lot of consensus here. Half the community wants bland flavorless piles of mechanics they can describe however they want and the other wants a separate unique class for swashbucklers, knights, samurai, gladiators, gishes, etc.

Part of the design issue is that the fighter alone doesn't mean anything: a fighter represents any adventurer who can swing a sword. The wizard represents a specific type of person: someone who learned a specific type of magic via a specific method. And too many people want the fighter to be that generic "fill in the blank" class for them to change it.

That's a 6e discussion. Cutting and elevating some subclasses and toning down some classes.

But when it comes to 5e and the wizard itself as a class and possibly the Warlock and cleric..

We might need to have a discussion about

Subclass specific spells.​

Because really there's certain things that people want to be able to have in some classes that really require a specific spell handcrafted around that specific experience that is not available to the base class.

If you want some ultraspecific fire spell or effect, it might require taking that specific fire based subclass which locks you out of access to other Elemental spells. And that's spelled might not fit on any main class at all.

 

[Chaltab]

Honestly I generally agree with 5E's take. Going back through older editions it always kind confuses me that certain thing were considered to warrant their own classes, like Samurai, Illusionist, Assassin, and Invoker.

There are some that I don't think a subclass cuts it for (Warlord and Shaman for instance) but still.

 

[Not a Decepticon]

Something about class and subclass design from latest Matt Colville video - he was descriving classes in Draw Steel and cited MCDM team's experience designing Illiger, Talent and WIldheart for 5e. Namely, that the decision of not having multiclassing in Draw Steel was made by how broken multiclassing is in 5e and how often they had to completely nerf othertwise awesome abilities for these classes, just because too many beta testers pointed out that through multiclassing you can get an utterly broken combo between it and something else.

 

[Remathilis]

Honestly I generally agree with 5E's take. Going back through older editions it always kind confuses me that certain thing were considered to warrant their own classes, like Samurai, Illusionist, Assassin, and Invoker.

There are some that I don't think a subclass cuts it for (Warlord and Shaman for instance) but still.

The line really depends on what you consider to warrant a full class or a subclass.

Psionics is a good example. The PHB has four "psionic" subclasses (a warrior, an expert and two different types of spellcaster) and for some, that's enough psionics for the game. Obviously, not everyone agrees as we just tested a base class psion with it's own subclasses and unique spells. Even further back, they tried a all-in-one psionic class with it's own system and everything. The general consensus though is that no one can agree exactly how best to represent the idea: for some a sorcerer with ten bonus spells is enough. Others want a full class. Others still want a whole different magic system.

 

[DEFCON 1]

The line really depends on what you consider to warrant a full class or a subclass.

Psionics is a good example. The PHB has four "psionic" subclasses (a warrior, an expert and two different types of spellcaster) and for some, that's enough psionics for the game. Obviously, not everyone agrees as we just tested a base class psion with it's own subclasses and unique spells. Even further back, they tried a all-in-one psionic class with it's own system and everything. The general consensus though is that no one can agree exactly how best to represent the idea: for some a sorcerer with ten bonus spells is enough. Others want a full class. Others still want a whole different magic system.

And this is exactly why there is no correct answer. Every player will have a different opinion on how many classes is the right amount and how subclasses are the right amount too. And never the twain shall meet. So the D&D designers just make ones they think will be enough to sell.

 

[Merlecory]

Something about class and subclass design from latest Matt Colville video - he was descriving classes in Draw Steel and cited MCDM team's experience designing Illiger, Talent and WIldheart for 5e. Namely, that the decision of not having multiclassing in Draw Steel was made by how broken multiclassing is in 5e and how often they had to completely nerf othertwise awesome abilities for these classes, just because too many beta testers pointed out that through multiclassing you can get an utterly broken combo between it and something else.

I know that you might not agree with this, but I Despise this reasoning. Multiclassing is an optional rule that states it is not balanced. Just make your class, stop freaking out aout how a small fraction of the player base who opted into an unbalanced rule will have an unbalanced experience!

 

[Minigiant]

I know that you might not agree with this, but I Despise this reasoning. Multiclassing is an optional rule that states it is not balanced. Just make your class, stop freaking out aout how a small fraction of the player base who opted into an unbalanced rule will have an unbalanced experience!

5e was designed around multi-classing being optional.

The choice in 2024 of making multi-classing a base rule was a poor one.

Instead they should have been multiple very optionals rules that tables could have chosen to get options from other classes and just sucked up that this would be slight or major imbalance

1. Multi-class
2. Dual class
3. Class dip feats
4. XP penalties
5. Ability score penalties

But it's too late for that.

In 2024 multi-classing is a base rule and people get access to subclasses at level 3 so we have to design sub classes with assumption that their level 3 ability cannot be too strong in the hands of another class with just a three level dip. Especially in post level 5 where it is very cheap to do so.

 

[Horwath]

i won't deny i've actually desired even tighter focuses on baseclass caster spell lists with bigger expanded spell lists for their subclasses, because honestly even with the current lists there's still a ton of overlap in spell selection.

with this much overlap, I do not think that we need more than ONE caster class.

Have every spell available, but spell known to level of 2014 sorcerer.
As is, you can get ANY spell, but you cannot have EVERY spell.

rest of mechanics can be filled with subclasses and feats.

want to play a healer? Take the Healer subclass, that is more or less life cleric.

want to RP a cleric, have Channel divinity a feat.
Take 1 or 2 options and number of uses is PB per Long rest, +1 use on Short rest.
2 options if you want mimic 3.5e two cleric domains.

want to be a shapeshifter:
take a feat:
PB usage per Long rest, +1 usage per Short rest.
take a form of Beast or plant with CR of 1/4 or lower.

 

[CreamCloud0]

with this much overlap, I do not think that we need more than ONE caster class.

okay i did say there's too much overlap, i didn't say there was that much overlap

Have every spell available, but spell known to level of 2014 sorcerer.
As is, you can get ANY spell, but you cannot have EVERY spell.

rest of mechanics can be filled with subclasses and feats.

want to play a healer? Take the Healer subclass, that is more or less life cleric.

want to RP a cleric, have Channel divinity a feat.
Take 1 or 2 options and number of uses is PB per Long rest, +1 use on Short rest.
2 options if you want mimic 3.5e two cleric domains.

want to be a shapeshifter:
take a feat:
PB usage per Long rest, +1 usage per Short rest.
take a form of Beast or plant with CR of 1/4 or lower.

this is just going way too far in my opinion, a lone caster class with feat clip-ons is not something to strive for IMO, it is good that there are separate chassis with secondary mechanics worked in, i'm just saying, fullcaster spell lists are very versatile, even the restrictive ones like sorcerer, so trim all of the fat and get right to the bone of what ought to be on each caster's base lists, all druids might have speak with animals and entangle, but leave conjure beasts and dominate beast to the beastmaster druid and plant growth and grasping vine to the nature druid.

it's the thing of 'if you didn't already know this was an [insert specialization] wizard, would you really be able to tell from their spell selection?'

 

[Minigiant]

with this much overlap, I do not think that we need more than ONE caster class.

Have every spell available, but spell known to level of 2014 sorcerer.
As is, you can get ANY spell, but you cannot have EVERY spell.

rest of mechanics can be filled with subclasses and feats.

want to play a healer? Take the Healer subclass, that is more or less life cleric.

want to RP a cleric, have Channel divinity a feat.
Take 1 or 2 options and number of uses is PB per Long rest, +1 use on Short rest.
2 options if you want mimic 3.5e two cleric domains.

want to be a shapeshifter:
take a feat:
PB usage per Long rest, +1 usage per Short rest.
take a form of Beast or plant with CR of 1/4 or lower.

That's how we got the unbalanced Class/subclass situation of clerics and wizard.

If one class can get over 50% of magic, you have to nerf it or magic to hell.

Understanding class to subclass importance is very important.