D&D 5E Time to ditch the old ability modifier array?

[Chris_Nightwing]

There's a lot of talk about removing ability modifiers from damage, or from attacks, or from combat altogether, and I get the feeling this is because of the potential disparity between optimised and spread-out characters, as well as the scale of said bonuses when compared to other factors that influence damage and attacks. I think the current array, of +1 per 2 ability points away from 10, was made to reflect the range seen in earlier editions, though only really in dexterity and constitution (for fighters); strength was much less broad, and the other abilities didn't behave similarly at all. I have seen alternative arrays proposed, so how about we stride bravely forward (or backward) to a realm of less variability between ability modifiers (though not necessarily between other factors based on abilities such as carrying capacity)?

As a first shot (maintaining the assumption of 3d6 rolls for stats) how about:

1-3 -3
4-6 -2
7-9 -1
10-11 +0
12-14 +1
15-17 +2
18-20 +3

And so on (or larger jumps, I don't care, it doesn't need to be regular in my mind). The highest you'll get to start with is +2 (ignoring the current terrible humans), and the ability limit might want to be higher, 21 for instance, unless there are things that having 19/20 get you, such as feats/spells/items. What do you think - any other tidy ideas?

 

[delericho]

I'd be in favour of this. Although I would generally favour replacing the current formula (mod = (score/2) - 5) with (mod = (score/3) - 3), giving:

1-2 -3
3-5 -2
6-8 -1
9-11 no mod
12-14 +1
15-17 +2
18-20 +3

The key reason for that is that a formula is easier to remember than a table, and because it's slightly more generous than the table.

 

[am181d]

Would you get the same effect if you just lowered starting ability scores? I'd prefer it if most characters started with a 16 (or lower!) in their key ability score.

 

[delericho]

Would you get the same effect if you just lowered starting ability scores?

Probably.

I'd prefer it if most characters started with a 16 (or lower!) in their key ability score.

Use 4d6-drop-lowest as-is, and you should get that effect, at least until racial adjustments are added. Likewise, if they did more to make all stats valuable to all characters, that should drive the optimisers towards more 'balanced' arrays, rather than the current drive to having one or two high stats and dumping everything else.

(As for how to make all stats valuable to all characters, well, that's the real trick. I would recommend tying at least one or two important (but not class-specific) things to every stat, making sure these come up in play, and not providing any means of substituting one stat for another. But that's all easier said than done.)

 

[Chris_Nightwing]

For sure more effort should be made to make every ability relevant to *something* used by each character - more than just skills.
I would have gone for a formula, but wanted 10-11 at the centre (since the mea of 3d6 is 10.5).
Another idea would be widening the brackets as you move further away from the mean of course..

 

[howandwhy99]

I'd prefer a series of arrays for different game elements, but as crazy complex as they get in other parts of the game I doubt they will diversify this element.

The 10.5 average matters, but only if characters are actually generated to some respect. This means rolling stats and picking stuff your character can pick like class and equipment (forget about race ). Gaming a character into development is a big deal now and for many an important part of the game, so that option may become the default instead. They've already dropped modifiers as representative of a bell curve, so I doubt it matters anywhere near as much as it did.

 

[Chris_Nightwing]

Hm, I had a thought over the weekend that maybe the problem with the ability modifier array is its linearity. What if instead we used triangular numbers to determine your modifiers? Yes, it's not a simple formula, but I think triangular numbers are great for providing differentiation between characters without incentivising power-play.

1-3 -3
4-7 -2
8-9 -1
10-11 0
12-13 +1
14-17 +2
18-23 +3
24-31 +4
32..

(This would be easier if we didn't use 3d6)

 

[Li Shenron]

Hm, I had a thought over the weekend that maybe the problem with the ability modifier array is its linearity. What if instead we used triangular numbers to determine your modifiers? Yes, it's not a simple formula, but I think triangular numbers are great for providing differentiation between characters without incentivising power-play.

I don't understand why it would make things better... All that really matters in the game is the modifier, because that's what applies to rolls. There might be some instances where the score is applied instead, but AFAIK they are always rare and not really different from using modifiers.

Linear or non-linear is a design choice that is related to (a) creating a certain wanted distribution of that score over the population of player characters and (b) setting a certain wanted rate of improvement by level, or temporary boost by spells and items.

For (a) it doesn't really matter whether you design the distribution in the score or in the modifier. With 3d6 rolled for the score and then linear relationship between scores and modifiers you get a certain distribution, but you could get substantially the same with flat score distribution (e.g. 1d16+2) and then some non-linear relationship between scores and modifiers that yields the same result. Therefore having 2 non-linearities here hardly helps.

For (b) you can always control the resulting modifiers directly, e.g. when designing spells and items. Anyway also when choosing the ability score increase by levels, the designers should always first think in terms of modifiers and then figure out how often those ability boosts should be granted.

Perhaps one indeed interesting utility of such non-linear, progressively slower relationship between scores and bonuses, could be that of making spells (and items) progressively less useful for characters with higher scores, and by consequence of higher levels. Thus, a +4 strength belt would be very useful to a weak PC, fairly nice to a strong PC at low level, and totally useless for a strong PC at high level (if the bonus to score doesn't trigger any bonus to modifier). It would make design much complicated tho, with possibly some weird side-effects that not every gaming group would be comfortable with...

 

[dkyle]

I'm in favor of having no ability score modifiers at all, low ability prereqs for classes, and just using a roll under system for resolving non-combat tasks (basically, skills).

I'm a 4E fan, but OD&D actually got abilities mostly right; ability scores should be more about describing your character in very loose, general terms, not applying bonuses all over the place. In particular, applying ability modifiers to both combat and non-combat bonuses in a highly combat-focused game like D&D encourages cookie-cutter builds where nearly everyone maximizes whatever stats are most useful in combat.