Wednesday, January 12, 2011

Thinking about game design: ability scores

Looking at EPT makes me think a bit more critically about ability scores. The classic D&D model of 3-18 scores with secondary modifiers means ability scores have limited purpose or value (the old school model in which extremes have a mechanical effect) or that they have a great deal of influence (3.0 edition on), but only because modifiers an the high scores that go with them are essential.

There are more subtle approaches. Systems like Fudge and Marvel Superheroes introduce some interesting mechanics around qualitative descriptive rankings. Some games (like Talislanta) use modifiers as scores. Storyteller systems use pips, etc. But conceptualizing what one dot of strength or feeble strength means can be challenging to explain.

EPT ranks abilities 1- 100, allowing them to show a range of numbers on a known scale and to serve as a simple task resolution system. Other games have done this since to differing degrees, but this still excites me. What's the chance of a STR 57 have of forcing a non-fortified door? 57%. Maybe a barred one we half that.

Hey, but there's more we can do. We can tie that score to quality of result. If you succeed, you succeed by an amount equal to the percent rolled. You have an 80% Strength and try to lift a portcullis to your waist. You roll 50 and you're halfway there. The guards get a little closer, and you have to succeed again....

I do have problems with the swinginess of flat percentile rolls to determine ability scores. I want some curve on that distribution, and firm limits on the minimum and maximum. There are also some problems I have with flat percentages since results are also too swingy in play, but I bet my Basic RolePlaying rules that this is more a problem in my head than in play.

Which all says that thanks to Mike saying he's thinking about making a generic EPT clone, I am back to thinking about what I can take from Dr. Barker's funky white box variant.

6 comments:

  1. Hi Doug, I posted an early draft on my blog today. I am with you regarding the flatness of 1d100 rolls (I had the same issue with Rolemaster). I (like you) also like the higher degree of variance 1-100 gives you for attributes however. The more I read the EPT rules, the more I like that version of D&D - though I am finding some problems with it too.

    I may post on them later as I organize my thoughts around class balance with regards to warriors...

    ReplyDelete
  2. Check out OpenD6... the current incarnation of the classic Star Wars D6 from West End Games. Attributes (ability scores) are the default scores for skills. And the system uses a dice pool so the probabilities are bell-curved rather than flat. Want to get smarter/stronger? Just spend your advancement points on your abilities.

    ReplyDelete
  3. In order to create some curve when it comes to the determination of ability scores, you could throw not two but ten d10 and add the ciphers, 0 counting as 10.

    ReplyDelete
  4. I've been playing around with the idea that your Statistics are your target numbers, and that you modify the role. For example, if you have a Deception Statistic or 11, then you want to roll that number on 2d10 or less, and that die roll is changed depending on the circumstances of the role (How difficult the circumstances, skill modifiers, and so on).

    Of course, the problem is that I've been trying to decide if I need an advancement reward system or not.

    ReplyDelete
  5. @Rerun941: thanks for the referral to openD6. Of course my personal brain fart with dice pool games like D6, Storyteller, or L5R is that I still can't easily parse the probability of an outcome in a dice pool. How much does adding a die increase the probability of success? What is my chance of succeeding with a pool of more than one die? And how do I answer those questions for players?

    @schlossblick: one of many good ways to get a bit of a curve back in (5d20 or 10d10 are good options). The old TSR game Gangbusters had modifiers to base d100 rolls to even things out a bit, and I remember Star Frontiers doing a cross-reference table (though that could be faulty memory). What I think might work for my head better would be something more like Warhammer FRP which has a set of d10s + a bonus to get a limited range with a minimum score and a maximum to have a curve within set limits.

    @kensan-oni: a very similar idea, and worth considering, but the probability issue of curved results on action resolution while more realistic, always drive me nuts in figuring out. The increase in outcomes for bonuses near the middle of the curve are much different than they are in the extremes. Not that that's necessarily a problem (in fact, it seems pretty realistic -- slight variances far from the mean have deep effects), but it hurts my brain to figure out how to read their effects in actual outcomes.

    ReplyDelete
  6. Oh, and Mike, I have downloaded the Swords of Abandon rules and will be giving them a look. My quick skim looked nice and familiar, though.

    ReplyDelete

Print