Ability Scores - Should they increase?

[Spatula]

And two-- it becomes much harder for DMs to guesstimate the numbers he needs to set up for encounters because they are just so big that it's hard to visualize. Most of us all know instinctually that at 1st level, ACs of 15 or 16 are average, an AC of 20 is really good, and a monster with an AC of 22 is a beast. But what are those numbers for level 15? Anyone know? I sure as heck don't (without pulling out the books and doing a lot of the math to get there.)

If we're talking 4e, an average AC for level 15 is 29 (level + 14). Average non-AC defense is 27. Average attack vs AC is +20 (level + 5), vs non-AC is +18. Those are monster numbers, but the PC numbers track pretty closely if you keep your ability scores, equipment, and feats up to date.

If we're talking 3e, it probably doesn't matter, at least not for PCs. At that level, monster attack bonuses far outstrip AC unless you're devoting all your resources to it.

After looking at the mathematical problems with 4e at the upper levels, I'm in the "ability increases are bad" camp. Also, I'd be down with reducing the impact of the ability bonuses. Castles & Crusades uses something like: 13-15: +1; 16-17: +2; 18: +3; which is a little less steep.

 

[CleanCutRogue]

Compromise. Offer increase at a very slow rate. +1 every 5 levels, or at levels 10 and 20 and that's it. Or allow attribute development at the expense of skill ranks or something, representing honing of natural talent.

 

[foolish_mortals]

erph, just get rid of them. Make it how it used to be, a special event. Make gauntlets of ogre power and belts of giant strength all the more sexier. The Wish spell was a good way to make points go up. Special occasions only.

foolish_mortals

 

[CleanCutRogue]

erph, just get rid of them. Make it how it used to be, a special event. Make gauntlets of ogre power and belts of giant strength all the more sexier. The Wish spell was a good way to make points go up. Special occasions only.

foolish_mortals

simple enough argument. You convinced me. I'm on your boat, captain

 

[Fanaelialae]

My preference would be for more gradual bonuses coupled with either a hard cap, or diminishing returns that effectively accomplish the same thing. Something like:

Score Modifier
1-4 -2
5-8 -1
9-12 +0
13-16 +1
17-20 +2
21-24 +3
25 +4

Regarding the above, I'd say set the general cap at 21, with penalized scores capped at 16, and scores with bonuses capped at 25. Racial bonuses / penalties would still be the normal +2 / -2 we've seen in recent editions, to leave room for improvement.

I also like the idea of an option rolled stat system, where lower bonuses pay off in the form of faster increases and even possibly a higher overall end bonus. This isn't without precedent, as there are numerous novels where the young and inexperienced kid ends up being the strongest in the end (someone who has great, but unrealized, potential).

 

[Viktyr Gehrig]

I like increasing ability scores. While it's true that no amount of training can make a human being capable of dead-lifting a ton... it's also true that no amount of training can make a human being capable of a thirty foot standing broad jump or swimming upstream in full plate armor. Those are also things I want my high-level D&D characters to do.

I dramatically increased the rate of ability score increases in my Pathfinder house rules: a +1 increase to three ability scores every 4th level, and a +1 increase to all of your ability scores at 6th, 10th, 18th, and so forth.

 

[Lanefan]

Stat increases should be very uncommon or extremely slow or both.

The percentile increment system that 1e Cavaliers had works well for all classes - you get to roll some dice each level-up (and IME people love rolling dice!) but you're only likely to have a stat or two go up by one point during your career due to it.

Rare and expensive stat-boosting items, both temporary and permanent, are also fine by me.

That said, the 3-18 bell curve lends itself so very well to the highly elegant roll-under-on-d20 ability check mechanic...which itself removes the need for about three-quarters of the 3e skill table and a bunch of other things. That right there is good enough reason to only allow stats to reach 20 in the most unusual of situations.

As for bonuses, I can easily live with a non-linear bonus set-up. I also don't mind a large range of middle (say, 7-14) numbers giving no bonus/penalty at all; this reflects that these individuals are at the middle of the bell curve.

What makes adventurers special is they get to roll (yes, roll) their attributes on 4d6 or even 5d6, as opposed to normal people who are stuck with 3d6.

Caveat: Strength is the one stat where this model falls down, as we need more granularity of bonuses over a larger range. 1e did this by adding percents to an 18 Strength; in my system this falls apart when trying to combine it with the Cavalier-like percentile increment system I refer to above. The answer I've found is to split out the 18.xx strengths into their own numbers but treat ability checks as is the Str is 18; but this is probably more complicated than people are looking for in 5e.

Lanefan

 

[Spatula]

Caveat: Strength is the one stat where this model falls down, as we need more granularity of bonuses over a larger range.

Why do we need that?

Personally, I was never a fan of roll-under ability checks. Skills are something that you can improve, at least.

 

[howandwhy99]

I think A.S.s should increase do to actions in the game. Read a tome, drink a potion, cast a spell, jump in a fountain, something like that.

Training in a stat for a long time I could understand working as well. What would balance that would be the time cost or some other cost, so it just didn't come free of charge.

 

[Lanefan]

Personally, I was never a fan of roll-under ability checks.

I like the sheer simplicity.

Skills are something that you can improve, at least.

Fair enough, but again at cost of simplicity. Playing 3e I found the biggest headaches at level-up were futzing around with skill incrementing and feat choices. Easier to abandon the global skill system except for a very few (ride, swim, non-Thief climb, boating leap to mind), then design something new that works for skill-based classes e.g. Thief.

Lanefan