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

[Merlecory]

Some other thoughts, now that I am not at work.

• Subclasses need to hit hard mechanically and thematically right away. An Arcane Archer and a Battle Master who uses a bow should feel different right away.
• Classes like rogue have a long gap between subclass features. I would say that this broadly doesn't work for subclasses. I would posit that the Arcane Trickster is the exception to this rule, entirely because they gain subclass related features on other levels. For this to work in other subclasses, I think both number of uses, and scope of benefits needs to grow. More uses of my 3rd level ability isn't as impactful (thematic power, not just mechanical power) as gaining new ways to express the subclass identity.
• I am interested in (but doubt WotC would ever make) subclasses that exchange base class features. Is there an interesting Barbarian subclass that loses the ability to ever reckless attack? It might be a more interesting thought experiment than subclass, but I don't know for certain.

I think there were Mike Mearls Happy hour streams where he talked about designing (sub)classes. Does anyone know if there are VODs?At the very least, I turned up this thread on the topic.

 

[Reynard]

.

• Classes like rogue have a long gap between subclass features. I would say that this broadly doesn't work for subclasses. I would posit that the Arcane Trickster is the exception to this rule, entirely because they gain subclass related features on other levels. For this to work in other subclasses, I think both number of uses, and scope of benefits needs to grow. More uses of my 3rd level ability isn't as impactful (thematic power, not just mechanical power) as gaining new ways to express the subclass identity.

Wait, isn't the rate of gaining subclass features standardized across classes?

 

[Chaltab]

For me the biggest pitfall of subclass design is that you can't try and balance a weak feature with a strong feature at a later level, or vice versa because there's no guarantee what level the campaign will go to.

I was just reading Tome of Heroes earlier and noticed the hilariously mismatched level 6 features of the Path of Hellfire Barbarian and the Path of Dragons Barbarians.

The former can use their reaction to deal PB Fire damage as a reaction to a creature that hits an ally with an attack. Which means 3 damage when you get it, never more than 6, and it's one of the most commonly resisted types.

Meanwhile the Path of Dragons Barbarian gets a Legendary Resistance. Yes, at level 6. And a second one at level 14, in addition to their level 14 feature! The level 14 feature of the Hellfire Barbarian is pretty good, but if a campaign's only going to level 10 then it's irrelevant.

 

[Chaltab]

Wait, isn't the rate of gaining subclass features standardized across classes?

No? That's been a common criticism of 5E from the start.

 

[Vael]

Wait, isn't the rate of gaining subclass features standardized across classes?

No, it was tried for an iteration of playtesting, but aside from moving all subclasses to start at level 3, Subclass progression has not changed. So classes that got their subclass at level 3 saw no change, those that got it at 1st or 2nd level have to wait until level 3.

 

[EzekielRaiden]

Personally, I thought the 5.0 Warlock "split subclass" concept had a lot of potential. It just wasn't necessary for the Warlock, and the new version is better for making pacts non-unique.

The Summoner class I've been chewing on for a few years now uses this model. At 1st level you choose the nature of your Visitant, the far-plane being that has forged a bond with you through the hidden constellations of planar influences. Then at 3rd level you choose how your Visitant actually manifests, by selecting a recondite Star Sign which is the focus for your Visitant's power.

For the Visitant type, I went with Umbral, Elemental, and Celestial. Umbral could be Far Realm entities, beings of living darkness, or being a that wish to remain unseen while they explore the strange sensations of the mortal plane. Elemental could be dragon-like, genie-like, giant-like, or more purified elemental powers that want to learn about this (to them) strange plane where elements mix naturally. And then Celestial is pure "Biblically accurate angel", Annunaki, heavenly but disturbing beings.

For the Signs, I had three: the Muses, the Protean, and the Chimera. Muses makes the Summoner support-leaning, Protean skill-leaning, and Chimaera offense-leaning.

Then Invocations become Evolutions, spontaneous adaptations your Visitant manifests as it becomes more comfortable existing on the mortal plane.

 

[Blue]

2. All subclasses should have same power budget. That will leave open option for universal subclasses or just dissolving subclasses into feats with level requirement that anyone can take.

I'd like to try to change your mind on this one. I think it's potentially more negative than positive.

The biggest is that it assumes that the base classes are all the same power so that base class+subclass=the same(ish). And I know what you're thinking, that that doesn't seem too much of an ask. But take a look at 2014 races. When they tried to have more powerful races there was no design space left over. That's why we had things like medium-sized pony-people instead of actual centaurs, and the huge squawk (pun intended, dangit) over always-flying races.

So you end up with the full caster being the bare minimum base class, and if the class has even more, like wildshape, that pushes the base class budget up even higher.

At this point we have all base classes, before adding subclasses, being very powerful -- full caster plus more level. Then we need to add subclasses.

If we keep subclass power low, they're mostly an afterthought compared to the class. It's not worth having this big division, and all future subclasses will be limited to this level of power, leading to issues just like we had with 2014 races when trying to introduce others.

If on the other hand we make subclass power high enough to matter, then we're adding one enough power that's like an extra half a full caster. (Not a half caster, half someone with 9th level slots because that's the power level of the classes). At this point we've got wild amounts of power creep.

And what's the benefit of this? You mention universal subclasses, but your first priority of balance kills that because there is no way to design that every subclass will synergize with every base class in the same amount. Universal subclasses are a non-starter with your first priority in place.

So what does it really buy us? And is that worth the cost of design shackles of forcing all base classes to have the same power budget, which is full caster+?

As we saw with ten years of expansions with the 2014 races, there's a real cost to trying to make everything the same. Allowing that the (base class+subclass) total all have the same power budget (gained at the same rate) gives you a lot more flexibility to do things a class that has a heavy amount of power into a pet, a wild shape, or something else based on subclass with a weaker base class, and an powerful base class with lighter subclasses.

Basically, it unduly limits our design space in ways that we have experienced as real and not dismissable for a spreadsheet-balancing bonus that brings little to the table for actual character creation and advancement during a campaign. Neither the player nor the DM will notice if Alice's character's power is 80% base class and 20% subclass and Bob's character's power, equal to Alice's in actual play and growing at the same rate, is 65% base class and 35% subclass.

EDIT: This assume subclasses still exist. Doing this all to subclasses so "some future edition or 3pp or my houserules might break them into feats" doesn't bring anything to the table for people playing by the rules now and therefore has zero weight. However actually breaking them up to feats is a whole different thing and this is not an argument against that. It has challenges but it's a nifty idea.

4. Since Concentration is such a limiting mechanic(and really good one for the most part), no fixed class feature should give usages of spells that require Concentration(yes, Hunter's mark, I'm talking about you).
IF that feature exist, then it either needs to have option to be used on something not requiring Concentration or giving option to remove Concentration tag with some power reducing effect if needed:
as in with War cleric, Fey wanderer ranger or Draconic sorcerer.

Eh, since I'm commenting, I've long thought they've underused Concentration by only using it (almost?) exclusively for spells. I could see Fighter or Monk stances, temporary Paladin auras, maybe a ranger wild-shape subclass, and all sorts of things that classes that aren't full combat casters could use it for.

So I think this an underused design space to be utilized, not one to be avoided. Though I agree with you about Hunter's Mark and would add the Paladin smites to the list.

 

[Remathilis]

The power budget between class and subclass is important. There needs to be enough in each to insert flavor and choose in each.

Full casters have too much power in Class.
Rangers and Fighters have too much power in subclass.

I don't consider this a bad thing though. Fighter itself means nothing and your subclass should really be defining the type of fighter you are (knight, arcane archer, gladiator, etc). Casters on the other hand have huge amounts of baked in flavor just by their spell selection alone. You can't reflavor a bard as a different type of caster without major revisions to the spell list, and 10 spells in a subclass ain't always enough.

I think most subclasses though provide too few features and most come too late to really feel impactful. Most classes should have 4 or 5 and they should be complete by level 15. The bard has too few, the rogue too late, the sorcerer might as well only have two. They really should have opted for the unified progression for subclasses, but water under the bridge i guess.

 

[Minigiant]

I don't consider this a bad thing though. Fighter itself means nothing and your subclass should really be defining the type of fighter you are (knight, arcane archer, gladiator, etc

It's bad because Fighter means nothing.

Fighter shouldn't mean nothing.

If half the power and most of the flavor are in the subclass and not the class.. the class doesn't matter and the subclass should be the class.

 

[Corinnguard]

If half the power and most of the flavor are in the subclass and not the class.. the class doesn't matter and the subclass should be the class.

This problem was addressed in Level Up when each of the classes was given features to cover each of the pillars in the game. They also became a little more customizable than their counterparts in 5e by having features that gave the player more than one option to choose from.

It's bad because Fighter means nothing.

Fighter shouldn't mean nothing.

Agreed. The Level Up Fighter has more going for it now. It has features that deal with Exploration and Social Interaction. They can become proficient in any two combat traditions at 1st level. So even before they pick up a subclass at 3rd level, they have started to become more distinct from one another. If a Fighter wants to become an Archer, they can pick up the Archery fighting style and the combat traditions, Biting Zephyr and Unerring Hawk. Then by 3rd level, they can pick up the Sharpshooter archetype.