Twist on the Skill+Attribute Dice Pool Mechanic

[Brad]

Many systems out there use a common Skill+Attribute system for determining the number of dice the player gets to allocate to his dice pool. The issue I have with this system is the importance of Attributes and Skill is always equal - it doesn't matter if you get 1 die from you attribute and 6 from your skill or vice-versa.

What I am wondering is if there is a way to diffrentiate between the two sources of possible successes? An idea that I haven't had the time to fully flesh out is that the different sources of dice use different types of dice. (ie. Attribute dice are d8 while Skill dice are d10).

So, instead of saying - "Roll a number of d10 equal to your Skill + Attribute." - you would say - "Roll a number of d8 equal to your Attribute and a number of d10 equal to your Skill."

The acutal die used in each instance could be variable. This could potentially be used to simulate situations in which Skill is far more important than the associated Attribute or vice versa.

The target number for both dice would be the same...making it easier to accomplish with one source and more difficult with another.

I think there is a potentially interesting mechanic here but I could use some feedback to flesh it out a bit....and poke holes in it.

[Ron Edwards]

Hi Brad,

The excellent game Swashbuckler does exactly this. Attributes are rated by die size (d6, d8, d10, d20), and they tend to ramp up to 3 d10's and 1 d20 after a few sessions. Skills are rated in units of d6's to add to the attribute roll.

There's been a lot of discussion of attribute + skill design on the Forge, and even before the Forge, going all the way back to the original Sorcerer mailing list, which you can find at its Sorcerer website. Take a peek at the early pages of the Theory forum too, and you'll find a couple of threads with titles about it.

Best,
Ron

[Mike Holmes]

I think that there are couple of others as well. Maddeningly, I am remembering commenting on one recently, and I can't think of what it is. I think that Matt Snyder's dreamspires uses something like this.

There are also a number of similar games that do other interesting things. Some games one part (either skill or attributes) are the number of dice rolled, and the other part is the die type. Or one part is dice, the other is target number for each die (or vice versa). Or one is dice pool, and the other caps successes. Interesting to note that there are often reverse versions of these. Often people cannot agree on which portion affects magnitude more, and which affects certainty (depends a lot on your definitions of skill and attributes).

In JAGS, skill simply defines the quality of the outcome, while the stats determine the magnitude.

In your system, Brad, if you could switch die type for both, how would the GM know when to switch. Just on gut feeling? Sounds like a difficult call. Also, are there limits on the total "sides" available? Like for 20, there would be the possibilities of d10 and d10, or d12 and d8 (or d4 and d16; saw em at GenCon). Or would certain circumstances leave the character better off than others in this regard?

Mike

[Matt Snyder]

I think that there are couple of others as well. Maddeningly, I am remembering commenting on one recently, and I can't think of what it is. I think that Matt Snyder's dreamspires uses something like this.

Some games one part (either skill or attributes) are the number of dice rolled, and the other part is the die type.

Mike, you did indeed comment on Dreamspire's Skill/Stat issue, but it's more like one you referred to in that last paragraph. That is, in Dreamspire, your attribute/stat is a number that indicates how many dice you roll, while your skill/talent is the type of dice you roll. Take the highest of the lot for your total.

[Brad]

In your system, Brad, if you could switch die type for both, how would the GM know when to switch. Just on gut feeling? Sounds like a difficult call. Also, are there limits on the total "sides" available? Like for 20, there would be the possibilities of d10 and d10, or d12 and d8 (or d4 and d16; saw em at GenCon). Or would certain circumstances leave the character better off than others in this regard?
Mike

Actually I hadn't though about limiting the total sides available. I was thinking more along the lines of using d10 as the default die and then reducing the die size in circumstances for which the Attribute or Skill is the dominating factor.

As an example, writing a computer program would be heavily dependant on the Computer Programming skill and raw Intelligence wouldn't help much. So, for this task you might roll a number of d10s equal to your Computer Programming Skill and a number of d4s equal to you Intelligence.

The target number for a success would be dependant on the complexity of the program... for the sake of argument lets call it a 6. So, for each Skill die you roll you have a 5/10 (50%) chance of a success.

Obviously, in order to allow for the possiblity of an unskilled character succeeding at a task, the die roll needs to be open-ended. In this case, the attribute dice would have a 1/4*3/4 = 3/16 (~19%) chance of success.

Determining which dice to use for a task is the tricky part and I haven't figured out a easy process for that yet.

[efindel]

I'd like to note that Skill + Attribute systems don't have to give equal weight to skills and attributes -- there's nothing that requires that the two have the same range.  For example, you could have a game where the "human range" for attributes is 0 to 10, with 5 being average, but skills range from 0 to 20, with 10 being "professional level".  In a setup like that, high attributes will help, but not nearly as much as high skill.

Ironclaw uses a system that I find interesting... attributes and skills are rated in dice, but those dice aren't added together -- you roll them and keep the best one.  With that system, gains made in the lower of the two don't expand the limit of what you can do (your highest possible roll), but they do make very low rolls less likely.

I've long had an idea of doing an attribute + skill system where one of the two is halved before adding in.  Normally, it would be the attribute, but in a few cases, it might be the skill.  The example that makes me want to do things that way is lifting stuff -- there, the strength attribute should be more important than any "lifting things" skill.

While I'm tossing out ideas, another I've seen discussed before is to have skill decide whether or not someone succeeds, and attribute factor into "how much" (i.e., into the degree of success).  For example, take two pilots, both with the same Piloting skill, but one with higher Dexterity.  Doing solo maneuvers, they'd both have the same chance of success, but the one with higher Dexterity would "do it better" with the same roll.  Where higher attribute would really come into play would be in opposed contests, where you don't just want to succeed, but want to succeed better than the other guy did.
--Travis

[Andrew Martin]

As an example, writing a computer program would be heavily dependant on the Computer Programming skill and raw Intelligence wouldn't help much. So, for this task you might roll a number of d10s equal to your Computer Programming Skill and a number of d4s equal to you Intelligence.

Just to be contrary...
I've been writing a lot of computer software lately, and just based on my own experience, it's really my own understanding of the problem that gets encoded into the computer program. My level of skill in Computer Programming is like my level of skill in walking, driving a car, writing skill, typing skill, carpentry or so on -- just an added capability. Which is why Miyamoto Mushashi can say that the art of a duel is just like carpentry...

[Le Joueur]

Hey!  Woah!

Please, let's not start the 'Skill + Stat for example X is more affected by [insert Skill or Stat here] according to [some named source]' argument!

This is game design, not reality emulation.  Sometimes there are very concrete reasons for structuring a mechanic along lines counter-intuitive (or unrealistic; as if there were any definitive science in the area of skill modelling).  Furthermore, this type of issue should be wholly the choice of the writer; there is no singular 'right answer.'

Please don't engage in pointless summation of your opinions upon how specific examples should be or how those examples should cause a mechanic to work; it is purely subjective in this field.  Your opinions do matter, but only so far as your design.  The topic of this thread isn't even about 'which is best,' so give the subjective back-and-forth a rest.

Fang Langford

(Who's just had a very hard night and apologizes for any apparent emotion in this post.  I really mean this in the best way.)

[Jeremy Cole]

Here's a mechanic that may get at what you're after.

You have a die level in all your attributes and all your skills.  All things are from D2 (best) to D20 (worst).  So, you may have Intelligence of D4 (real smart) and Computer Programming of D10 (average or mediocre).  

TO succeed at a test you must roll a 1 on the given die (so a D4 is better as its 25%, compared to 10% for the D10).

For a given test the GM assigns the difficulty, by telling you how many dice to roll, for BOTH the skill and attribute.  The skill (COMP PROG) is important so roll 3 dice for that, the attribute (INT) is not too important so roll only one die for that.  This allows the GM to give a weighting between attributes and skills.

For the above example, the player would roll 3D10s and a D4, and count up the 1s.

If the task was attribute heavy, you may roll the opposite, 3 attribute dice and 3 skill dice (3D4s and one D10).  Here the smarter, less trained character is at advantage.

Is this what you were originally after Brad?