Resolution mechanic I'm fiddling with...

[Colin the Riot]

SO, this is something I'm thinking of, and it's a combination of some things I've come across that I like.  Namely, Donjon (by Clinton), Clockworx (by Jared) and Shadows (by Zak).

Here's the thing.  I like in Shadows how you use the same roll for every conflict.  You roll two dice, one representing success, one representing complications.  Don't like how there is no definition to the characters inherent in the system.

I like in Donjon how successes equal facts that the player states which the GM narrates into the story.  Don't like dice pools (mainly because I don't have bunches of dice).

I like in Clockworx how if you have a skill, it allows you to manipulate the player's number towards the target number.  Don't like how it's a guessing game.

I have fudged those three into this:

In this game, characters are made up of descriptors that have numerical values 1-6.

When a roll is called for, the player rolls 2d6 vs. the GM's 2d6.  Hig roll wins.  The number that you win by equals the number of facts you get to state for the outcome.  Whoever loses narrates the outcome, but has to incorporate the stated facts.  GM has final arbitration.

The way skills come in is that if you have a relevant skill, you can boost your roll by the number in your skill, or reduce the GM's roll by that number.  Or both.  It would depend entirely on what one had in mind for the narrative outcome.

Example:  Dave's character has the descriptor "Sneaky(3)".  He wants to creep past a guard.  The GM calls for a roll.  Dave rolls a 6 and the GM rolls an 8.  Normally, the GM would win with 2 facts to use to make dave narrate a failure.  Something like, "the guard hears you", and "the guard has an itchy trigger finger".  However, Dave will use his Sneaky descriptor to boost his roll to a 9, giving him one fact to throw at the GM, like "the guard is asleep".

What sort of flaws might there be with this mechanic.  I'm not very good at math, but I know that with 2d6, you'll roll mostly 7s, so the descriptors really will matter in boosting your chances to succeed.

Please expose the flaws in this so that I might tweak it a bit.

[Jasper]

Well, it seems like a decent mechanic at heart, though two things jump out at me.  First, you say that when using a skill you can reduce the GM's roll, or increase your own, or some combination of the two.  Mathematically, this doesn't matter, since you're only looking at the difference, right?  Unless making the other guy's roll be less than zero does something special....

Secondly, sometimes you'll be having a lot of facts to narrate.   A roll difference of even four will be pretty common (standard deviation on each person's roll is 2.4).  If someone adds in a skill as well (and why wouldn't they?), it pushes it even higher.  Now, this may not be a problem, but it seems a little awkward to have to name ten facts say, yet have no narrative control over how they're strung together.  I can see you were trying to avoid the more standard "winner narrates" position, but maybe it makes sense?

Perhaps the winner could narrate, but the loser gets to incorporate a fact or two into it, depending on how muhc he lost by?

[damion]

Sounds ok, I'd restrict descritpors a bit more. A 6 descriptor would almost ensure  success, even if penalties worked like skills for the GM.

You could use, say 2D8,2D10, ect to get a bigger spread of results.

Search an Handeling Time: This may not matter to you, but I'm not a bit fan of universsaly opposed rolls.  The reason is the GM has to be involved in every roll, thus you can't have multiple people roll what their doing at once.
However if rolls are fairly uncommon,  only a few per scene, this is ok.
Another solution is in This thread.

[Stuart DJ Purdie]

sometimes you'll be having a lot of facts to narrate. A roll difference of even four will be pretty common (standard deviation on each person's roll is 2.4).

Wouldn't the easy solution to that be to say that you get a fact for each mukltiple of two (or three, or whatever) you exceed the other roll by?  In other words, just scale the number of facts down.

Division by two would make the normal case one or two facts, which strikes me as about right, and limits the maximum to five (eight with skills included), occuring once every 1296 rolls.

The distrubution for |2d6 - 2d6| (which is what your effectivly doing) is:

Code Select Expand


Value  Prob         / 2          / 3
0:     22.530       22.530       22.530
1:     21.604       40.894       56.942
2:     19.290       28.392       26.384
3:     16.048       14.040       5.244
4:     12.344       2.628        0.154
5:      8.640       0.770
6:      5.400    
7:      3.086      
8:      1.542    
9:      0.616    
10:     0.154    
Mean    2.74        1.54         1.26


As you can see, either with or without a divisor, 22% of the time youre going to get net 0 facts (barring skills).  You might want to consider what that will mean, in terms of narrating the outcome.  The columns head /2 and /3 are for division of the difference.

Stuart
In a mathematical mood...

[Colin the Riot]

To clarify a bit:

Jasper, when I say you can use a skill to boost your roll or lower the GMs, I know it's the same mathematically, but the narrative explanation would be different.  And I think reducing the other guy's roll below zero would be a "critical" which would allow you your facts AND allow you to narrate them.

I agree that a lot of facts will be going back and forth.  Perhaps placing a cap on the number of facts?  Like say, 6 maximum?

Damion, for the descriptors, a 6 is almost godlike in ability and is the absolute highest rating you can have.  So the fact that that would ensure success is sort of the point.

As far as the universally opposed rolls, I only do that because I like to roll dice as the GM.  I suppose target numbers would be just as easy.  I intend for rolls to be used only when a conflict arises wherin a chance for failure would enhance the story, so depending on the situations that the characters find themselves in, there may be more or less rolls.  As for people rolling their actions simultaneously, I don't see that being a problem.  If they are all involved in the same conflict (same obstacle) then the GM would only roll once.  

Stuart, perhaps this method of dividing would be a better way to limit the successes rather than just putting a cap on them.  As for the 22% netting 0 facts, I assume that means the GM and player would roll a tie.  That's why having applicable descriptors are important.  

I'm thinking in cases of a tie, a 3rd d6 would be rolled for each side to break it.

[Ron Edwards]

Hi Colin,

Interesting thread. I like talking about this stuff.

I'm curious, though, about what sort of RPG context this system works for, in your estimation. I guess I'm looking for character, situation, that sort of thing, very briefly.

What would be going on, imaginatively speaking, such that this system is enhancing the fun?

Best,
Ron

[Colin the Riot]

Ron,

The type of game I'm thinking of using this with is the sort of game you'd run with Zak's "Shadows".  However, rather than having the player state two possible outcomes, I'd like to use the mechanic in Donjon where you win with "facts" you are able to insert into the story.

I also would like for there to be some sort of descriptor system that would influence this mechanic.  These descriptors could be anything, and as long as the player could give a plausible explanation for using that descriptor in a given roll, he could do so.

The type of role-playing I see this enhancing is a game where the focus is on narrative tension.  The GM would have a loose story structure in mind, and the GM and the players would work together to reach the end of the story, using the resolution mechanic whenever there is opportunity for interesting conflict.

I'm not married to the particular mechanic that I've come up with.  I'm primarily trying to think of a way to use Donjon's "successes=facts" idea without using a dice pool.  Any suggestion on how I might be able to do that other than what I've got going already?

[contracycle]

I've been teasing out this idea a bit and I like it lots.  It has some interetsing implications so I thought I'd pick up this point.

I'm curious, though, about what sort of RPG context this system works for, in your estimation. I guess I'm looking for character, situation, that sort of thing, very briefly.

I'm not sure it needs a specific one more than say HW - look at what it does to protagonisim.

Say by a slip of the dice, a NPC kills a PC.  Even if the GM is harsh enough to hand out facts which include having three foot of steel through the gut, the player as the loser of the roll gets to narrate their own death-scene.  At the very least this allows the players natural action to be last words or a symbolic gesture or even potentially a post humous triumph, all becuase this power of realisation is located in the players action.  In conventional designs, the receiver of a fatal blow is the inactive participant - the circumstances of the characters demise are down to the active characters success or failure.  So frex, a PC might be cut down by the GM's combat monster in action one, and their death gets no spotlight.  The spotlight is still on the ocvmbat monster say resolving its next attack.  This allows MUCH more potential for a characters exit to occur in a positive and potentially still protagonising manner.

All this by way of simply extending the effect of this, umm, asynchronous mechanic; becuase the inputs from each side interact both actual people are actively contributing, one is not a recipient of the others simple fiat.

Some other interesting interactions - multiple parties to a conflict.  The order of action will be very important becuase the narration will be "handed off" from one party to another as decisions are resolved, and the thread may well switch back and forth between sides more than once.  It also potentially resolves the kung fu master vs mooks thing by making an opportunity for a mook to fall over occur every time they fail a role rather than every time the KFM succeeeds at one - hence he doesn't need a bajillion attacks.  Thus, the freeze frame combat problem goes away becuase even when its the mooks making ten attacks, the KFM "wriggles" through the manipulation of facts they feed back into the system.  

The only down side I see to this is that mandating facts might blur in=to de facto narration which it may be hard for the other party to reintertpret.  But I think it offers a very interesting perspective on the ordering of resolutions and who gains power to realise them.

[contracycle]

And I think reducing the other guy's roll below zero would be a "critical" which would allow you your facts AND allow you to narrate them.

Ah good - that occurred to me woth the KFM thing, the KFM should be able to control the movement of others - throws and stuff - when they succeed particularly well.

[Colin the Riot]

Okay,

I've been thinking some more about this.  I'm thinking I want to remove the descriptor mechanic altogether.  Right now, I'm more into the way this reolution system engages the players independent of characters.  With that in mind, how about this?

If you win the roll, the maximum amount of facts you can state is five.
Using something similar, again, to Shadows, each player would be given say, 5 tokens (including the GM).  These tokens are what a player would use to manipulate die rolls.  By spending a token, you can boost your roll one point.  

The way in which you would regain tokens is by earning facts but instead of stating them yourself, you would turn them over to the GM.  This would gain you 1 token per fact.

If there are few people playing, each person could actually get by with 1d6.  If too many people are playing at a time, 2d6 could be used, and the difference between the player's and GM's roll would be divided by 2 to get the number of facts.  If even more people are playing, 3d6 and divide the difference by 3, and so on.

Now, I'm still working out this token idea, but could this work?  Maybe players could steal a fact from another player's roll by giving them a token?  I think it might be neat if the players were somehow spending this currency amongst themselves.

Besides the addition of this token idea, the roll would still work the same.  Player vs. GM, higher roll wins, with the difference representing facts the winner gets to state about whatever the roll was called for.  Loser then narrates using the facts provided by the winner.  GM has final say.

Thougts?

[Colin the Riot]

Don't want this to drop too low.  I'd like to hear from Ron and contracycle again.