D&D 5E Class-based Ability Score Increases (revised proposal)

[clearstream]

[EDIT] After starting this thread, I realised what my motivations were as a designer. Looking at floating stat bumps, increased arrays, or extra dice rolls: to me they just feel bland. Racial ability score increase traits are flavourful, and I believe the most successful design would replace them with something equally flavourful. Could class-based increases along the lines I explore below evolve into that flavourful successor? What would need to change, to make it so?

[EDIT 2] Tweaked to make the connection with class stronger!!

Non-stereotyped​

All racial Ability Score Increase traits are removed.

Class Ability Score Increase features​

The class you choose at 1st-level gains a new feature.

Artificer Training (C)​

Increase your Intelligence or Constitution ability score by 2, to a maximum of 20.
Increase any one other ability score by 1, to a maximum of 20.

Barbarian Training (C++ MAD)​

Increase your Strength, Dexterity or Constitution ability score by 1, to a maximum of 20.
Increase any two ability scores by 1, to a maximum of 20.

Bard Training (S)​

Increase your Charisma ability score by 2, to a maximum of 20.
Increase any one other ability score by 1, to a maximum of 20.

Cleric Training (A++)​

Increase your Wisdom ability score by 2, to a maximum of 20.
Increase any one other ability score by 1, to a maximum of 20.

Druid Training (A)​

Increase your Wisdom ability score by 2, to a maximum of 20.
Increase any one other ability score by 1, to a maximum of 20.

Fighter Training (B+)​

Increase your Strength or Dexterity ability score by 2, to a maximum of 20.
Increase any one other ability score by 1, to a maximum of 20.

Monk Training (D MAD)​

Increase any one ability score by 1, to a maximum of 20.
Increase any two ability scores by 1, to a maximum of 20.

Paladin Training (A+ MAD)​

Increase your Strength or Charisma ability score by 1, to a maximum of 20.
Increase any two ability scores by 1, to a maximum of 20.

Ranger Training (C+)​

Increase your Strength, Dexterity or Wisdom ability score by 1, to a maximum of 20.
Increase any two ability scores by 1, to a maximum of 20.

Rogue Training (B)​

Increase any one ability score by 1, to a maximum of 20.
Increase any two ability scores by 1, to a maximum of 20.

Sorcerer Training (B++)​

Increase your Constitution or Charisma ability score by 2, to a maximum of 20.
Increase any one other ability score by 1, to a maximum of 20.

Warlock Training (B+++)​

Increase your Charisma ability score by 2, to a maximum of 20.
Increase any one other ability score by 1, to a maximum of 20.

Wizard Training (S+)​

Increase your Intelligence ability score by 2, to a maximum of 20.
Increase any one other ability score by 1, to a maximum of 20.

Multiclassing​

When you gain a level in a class other than your first, you don't gain the 'Training' feature of that class.

 

[DND_Reborn]

So, in a nut shell, you have two floating ASI +1s:

General Training (all classes)​

Increase any two ability scores by 1, to a maximum of 20.

And another ASI +1 based on a choices of ability score options determined by class selection?

It's perfectly fine, of course, but just not my thing. I am in the camp of just removing ASI bumps for race and/or class completely as I don't feel they are needed.

 

[Warforged DK]

I like the idea of class based Ability increases, and I've typically ruled them as +2 in your primary attack stat and +1 to any other. I've also done it tied to Saves, but that doesn't work out quite right.
My thoughts behind it were that the racial features were what made them who they are. Dwarven resilience would be the species trait and stonecunning is more of a cultural trait tat sets them apart. It's not a +2 in CON that does it.
Also, I don't understand your tier grading or why it's even necessary.

 

[clearstream]

So, in a nut shell, you have two floating ASI +1s:


And another ASI +1 based on a choices of ability score options determined by class selection?

It's perfectly fine, of course, but just not my thing. I am in the camp of just removing ASI bumps for race and/or class completely as I don't feel they are needed.

I've noticed that for nearly every (something like 95%) character generated in the groups I've played with, players choose their race with their desired class in mind. Additionally, they seem to enjoy being able to feel strong in that class, and where necessary being able to make their character more suitable (for example, where they have an odd ability score, it can just feel good to make that even and gain the bonus).

Does 5e need it? In terms of baseline balance, overshadowing, and challenge, that very much depends what character generation system a group chooses.

 

[clearstream]

I like the idea of class based Ability increases, and I've typically ruled them as +2 in your primary attack stat and +1 to any other. I've also done it tied to Saves, but that doesn't work out quite right.
My thoughts behind it were that the racial features were what made them who they are. Dwarven resilience would be the species trait and stonecunning is more of a cultural trait tat sets them apart. It's not a +2 in CON that does it.
Also, I don't understand your tier grading or why it's even necessary.

The tier grading and consideration of MAD classes has a significant outcome. Say I mandated instead +2/+1? That will work nicely for many standard class stat arrays - where the class benefits from a single high score - but it can be less helpful for MAD arrays. Monk is a good example.

I then also gave more freedom to classes where a perfect array is less likely to overshadow or press a DM into bending encounters around them. That concern comes out of acres of discussion and analysis on these forums where certain arrays in conjunction with optimal picks strain the system in various directions.

Regarding Mountain Dwarfs, I believe they are the only +2/+2 core race because they are roughly half an ASI behind their kin and Strength isn't such a valuable ability in 5e. Thus I think they need about a half ASI in balance. Or not. As you can see, I didn't feel that races necessarily needed compensating!

 

[clearstream]

So, in a nut shell, you have two floating ASI +1s:
And another ASI +1 based on a choices of ability score options determined by class selection?

Oh, another note. That construction is used because players will often want to be able to push a primary stat up +2, but equally one wants to generally forestall +3. The class selection is also expected to guide novice players: it signals where they will want to focus (for groups that allow some form of assign-as-desired in character generation e.g. every AL group!)

 

[DND_Reborn]

I've noticed that for nearly every (something like 95%) character generated in the groups I've played with, players choose their race with their desired class in mind.

I agree with that. In fact, yesterday when a player was contemplating a new PC, his first thought was what race he wanted to play for a concept. I commented how different it was, because (for myself, too) IME I see a choice of class as primary, and then race as secondary (i.e. "What race best supports the class I want to play and the concept I have for it?"). I understand why people want to remove that link, because it leads to players choosing races with ASI bumps where they feel they need to "feel strong".

Additionally, they seem to enjoy being able to feel strong in that class, and where necessary being able to make their character more suitable (for example, where they have an odd ability score, it can just feel good to make that even and gain the bonus).

Which leads to the next part, about feeling strong. While that is true, it is already built into the game without needing ASIs. Whether you roll ability scores, use the standard array, or do point-buy, you have higher scores to place--so place them where you want to "feel strong".

Especially if you use point-buy (the preferred method IME anyway), you can have all even scores so that isn't a factor either.

Does 5e need it? In terms of baseline balance, overshadowing, and challenge, that very much depends what character generation system a group chooses.

I think it just depends more on player viewpoint and the challenge level of the game the DM sets. Does a +4 modifier make that much of a difference over a +3, or a +3 over a +2? IMO it doesn't (from a player viewpoint or as the DM) but if it matters to other groups/tables that is fine, of course.

Other options are simple enough: increase point-buy points, bump up the standard array, roll 4d6-L seven times and keep the best six, arranging to taste. All of these options basically yield the same result as granting ASI bumps.

Anyway, I don't want to derail your thread further. As I said, I think what you're doing is fine, I just personally have never felt the need. Good luck with it.

 

[DND_Reborn]

Oh, another note. That construction is used because players will often want to be able to push a primary stat up +2, but equally one wants to generally forestall +3. The class selection is also expected to guide novice players: it signals where they will want to focus (for groups that allow some form of assign-as-desired in character generation e.g. every AL group!)

Sure, most alternatives work to grant a +2 and +1 bumps to different ability scores.

 

[loverdrive]

I think it's a bit overcomplicated. Something like "Distribute 3 points between ability scores, to a maximum of 2" is gonna do basically the same thing, but with less words.

 

[NotAYakk]

Each +1 seems small, but fall behind 4 and play starts feeling very different.

And that +1 was 1/4 of that gap.

When your friend hits on a 12+, almost half the time, you hit in a 16+, 1/4 of the time. In a 4 round fight with 8 attacks, 2 vs 3.6 attacks. In a slightly bad luck combat (1 SD) it becomes 0.8 vs 2.2; a 0 hit fight is 1.6 SD away for the lower attacker; 1.6 SD for the higher is 1.3 hits for the higher attack character.

Accumulated effects of accuracy is nuts.

This ignored secondary benefits as well.