Quote from: Dr_AvalancheAnother thing that I can't get out of my mind is that the task resolution should be based on skill, but the conflict resolution should be based on something else - something like Story Weight, for lack of a better word. I'm stumped how to do it elegantly though.
 

You can do this with metagame mechanics.  In Buffy for instance (you knew I was going there sooner or later ), you have the Drama Points that can be spent and rewarded for conflicts like this.  So in your example, there's the four possibilities.

1) The player makes his Stealth roll to sneak past the guard.  He gets by no problem.
2) The player fails his Stealth roll.  The guard spots him.
3) The player decides he wants something to distract the guard.  He spends a Drama Point and slips past while the guard is looking for the source of the noise - a cat or whatever.
4) The GM decides that he doesn't want the PC slipping past this guard.  He says that a maid walks around the corner and cries out in suprise alerting the guard.  The player is awarded a Drama point for being screwed over by the GM.

It works really well, and players don't seem to mind a bit of constructive railroading so long as they're getting payback.  They can use that DP against you later, it's one of the most important resources in the game.  Just don't overdo it, and remember to do it because it will make things more interesting, not to get them to do things "right".

Good stuff here. I need to run (work, work, work), but I just wanted to say that. This is stuff worth thinking about.
 
The cool thing is, it looks like it could be very easily implemented in a lot of already existing games without too much effort.
 
Edit: I was replying to kryyst, but it applies to Maddman's post as well...

Eh, I think that the action point idea would work, but is inelegent.  And, while Maddman the DM gives out something for arbitrarily screwing over the players, he's *still* arbitrarily screwing over the players.

No, I think that this idea deserves more exploration as a single, unified mechanic.  I don't really see it working in d20 -- any type of levels of success don't really work in d20 -- but I think it should be possible to do it with some sort of mechanic.

I just love the idea of DMs being forced to let SOMEBODY SOMETIME succeed at a stealth roll too much to give up on this idea.  

Quote from: GunhildaEh, I think that the action point idea would work, but is inelegent.  And, while Maddman the DM gives out something for arbitrarily screwing over the players, he's *still* arbitrarily screwing over the players.

Well the elegance comes in what else the mechanic is used for.  It's great for nudging play toward the genre conventions you want to go for.  They work really well, though to be honest I've only done this a couple of times.  I'm more likely to have the villian use their DPs to cause something to happen, such as the cops showing up to question PCs just as the demon runs away.

QuoteI just love the idea of DMs being forced to let SOMEBODY SOMETIME succeed at a stealth roll too much to give up on this idea.  

That is very much the feel, that someone spends a point they get to be GM for a second and say what happens.

Quote from: MaddmanWell the elegance comes in what else the mechanic is used for.  It's great for nudging play toward the genre conventions you want to go for.  They work really well, though to be honest I've only done this a couple of times.  I'm more likely to have the villian use their DPs to cause something to happen, such as the cops showing up to question PCs just as the demon runs away.

I am in no way opposed to action points.  I think they're a fine idea -- though I think Exalted's stunts might be a little cooler way of doing the same thing.

They don't, however, have the same feeling or inventiveness as Dr. Avalanche's initial idea.  That would be, as far as I know, something new under the sun for RPGs.  I'd rather explore the idea of a shiny new mechanic than cobble together a somewhat similar effect with old and moldy parts.  

Quote from: GunhildaThey don't, however, have the same feeling or inventiveness as Dr. Avalanche's initial idea.  That would be, as far as I know, something new under the sun for RPGs.  I'd rather explore the idea of a shiny new mechanic than cobble together a somewhat similar effect with old and moldy parts.  

Wait, are you saying that Unisystem doesn't do everything?

Actually, I like Dr. A's idea of it being part of the roll.  That way, it's not always under the player's control.

Hrm. I had an idea that I think is neat. Dice pools!

What, someone has done dice pools before?

Darn.

Ok, so that wasn't exactly the point.

:ponder:

I came up with an idea , which I'll present below, but the numbers are not worked on - I just inserted numbers that felt good. Ideally, I think characters should pass these tests most of the time, but far from every time - after all, an important part of these challenges are that the possible complications that come from failing are as interesting as succeeding, if not more so...

Anyway, here's what I came up with:

Every challenge you roll dice for, you use two kinds of dice - black and white (you can pick other colors, but I'll stick to the traditional, thank you very much).
Black represents your skill. It's a value that's written down on your character sheet in the old-fashioned way. In this case, I'm imagining a value around 1-5 or so, but that's something that could be fiddled with to get just right.
The white dice represents story power. You get one die for free, because regardless if this task is related to your story or not, you and the GM together has deemed it interesting enough to stop and roll for. In addition to that, you can get bonus dice for relevant traits on your character sheet. This could be various things - you might have a character motivation that is Greed, and you're doing this all to get to the crown jewels. You get a free white die for that. Another might be that you want to foil the king who imprisoned your brother. Since you bought that as another character motivation, you get another die in the white dice pool.
Finally, and this I think is the neat part, you really want to get past this challenge for whatever reason - the prospect of dealing with the guards seem particularly unappealing to you, for example - so you put yourself in debt to the GM, for a fourth white die.

Finally then, you roll your two dice pools. Every die that come up "positive" is a success. Each pool that reaches two successes passed the challenge - which leads to the four possible results in my original post.

- Skill success (black dice pool passed the test), Goal success (white pool passed)
- Skill success, Goal failure
- Skill failure, Goal success
- Skill failure, Goal failure

Personally, I think it looks darn neat... Obviously, unlike the above ideas, this would pretty much require a completely new game. I can't see anyone being able to insert this in any existing rules system...

Edit: Mixed up black and white dice in their final mention... Bah. Fixed now.

Heh. I was so excited about my new system, that I forgot to expand on one neat little finesse that I even bolded because I thought it was so cool.

By loaning dice from the GM, the player implicitly tells the GM "I really want to pass this, you can screw me over later in return, ok?" because the dice the player borrows, the GM can use to raise the difficulty for the PC next time he runs into a challenge.

Which of course is like the Uni-system GM giving the player Drama points in return for screwing him, but the other way around... "Here, have some dice, screw me over".

Edit: I started this before Dr. Avalanche posted again.  So this is just my own ramblings.  Now I need to go read his stuff.  


The question then becomes -- what sort of die mechanic *do* you use for this?  You need to have at least four possible outcomes.

With a d20, you could have a target number, with different results depending on how close to the target number you are.  But that requires a lot of on-the-fly math, which slows things down.

With a White Wolf dice pool, it might be a little easier.  In Exalted, a 7+ on a d20 is a success, with a 10 being 2 successes.  If you fail, and you have one or more 1's, you get a botch (a bad failure).

Maybe 1's could instead be complications.  For every 1 you roll, even if you succeed, your life gets more complicated.  Say you made your stealth roll, but you rolled three 1's.  The DM could decide then that the maid comes out and, with three 1's, is ready to scream her head off immediately.  You might need twice as many successes (6) on a bluff/intimidate/seduction/whatever roll to keep her from immediately screaming out an alarm.

Looking back at our initial goals:

Quote- The player succeeds with his action, and achieves his goal.
- The player succeeds with his action, but through circumstances, fail to achieve his goal.
- The player fails with his action but still achieve his goal through circumstances.
- The player fail with his action and fail to achieve his goal.

One and four are covered by all systems.  My idea opens up the possibility for number two -- you have an added complication, and you may or may not fail.

The problems: it doesn't cover number three.  More importantly, though, is that complications become likelier the more skilled you are.  It's the same problem that White Wolf botches have always had -- the uberbadasses are the ones the *most* likely to fuck up dramatically.  For many games, that would be just fine for everyone concerned.  But it just seems wrong that more skill = more complications and fuckups.


So that's my first stab at this.  Somebody else take a poke at it!  

Okay, now I've read Dr. Avalanche's posts.  It's rough, but I like it.  And I think it might be necessary for the idea.  If you've got two independant factors going on, two sets of dice seems logical.

And with two sets of dice, you could even have degrees of success or failure for both, leading to an even more robust mechanic.

And, while I think you'd have to have a new game system, I think it could be based on an existing one.  I could see a White Wolf-stylez game having this kind of die mechanic.  You'd just need to expand on motivations and/or virtues.

Quote from: GunhildaAnd with two sets of dice, you could even have degrees of success or failure for both, leading to an even more robust mechanic.

Very true. As long as it's intuitive - I think it's fairly important that the player and the GM together are able to define what success and failure means before the dice are rolled. If you create too much granularity, it becomes difficult to decide what the result actually means.

Of course, granularity might only be needed sometimes - in combat, for example. It could decide both who wins a fight, who achieved their objectives, and how injured they are.
I'm also wondering if this would benefit from open-ended dice, so if we're assuming d6's, each 6 would give both a success and a re-roll. That way, small dice pools wouldn't feel quite so discouraging. Not that the dice pools have to be small, but it seems more elegant somehow than rolling 20 dice for no particularly good reason.

I like it.  Simple and intuitive, and it leads directly to a more detailed narrative than a simple pass/fail system or even a success levels system.  One issue I see right away is that the goal pool is far, far more important than the skill pool.  After all, if the goal pool fails then you fail even if the skill roll succeeds.  And if the goal pool succeeds then then you succeed, even if the skill roll fails.  You may consider this a feature and not a bug but it leads to the obvious question - why have skills other than flavor?  If you can make sure you have a good goal pool then it doesn't matter what skills your character has.

Maybe not every roll could call for a Goal pool?  Or maybe have the skill pool dictate narrative power?  If the goal succeeds then the goal is achieved, but if the skill is also achieved then the player gets to determine exactly how that happens.  If the goal succeeds but the skill doesn't, then the GM decides what happens.  Conversely if you get a success with the skill pool but fail the goal pool the action fails, but you get to dictate how it fails.  If the skill fails then the GM ownZ j00.    This may have all been intended in Dr A's post, but it wasn't obvious to me at first.  It almost seems a little backwards, but still very cool.

Edit - thinking about how this would work in a fight.  The Goal is to harm or gain advantage over your opponent, the skill is your ability to use weapons.  And I'm inventing some nomenclature so we can discuss this.  The results of a roll will be denoted s and g together.  Capital for a success, lower case for a failure - SG, sG, Sg, and sg.  Not perfect but it will save me typing.

For SG it's pretty obvious - you hit your opponent.  For sG you don't hit them, but you do gain some advantage over them.  Perhaps they lose ground, get disarmed, or lose initiative next round automatically.  If you buy my BS above, the GM gets to determine this.  For Sg the weapon strikes true but doesn't wound or hinder your foe for whatever reason the GM dreams up.  A parry, doesn't penetrate armor, or a dodge.  And for sg you miss and likely place yourself at a disadvantage, GM's call.

Quote from: Dr_AvalancheVery true. As long as it's intuitive - I think it's fairly important that the player and the GM together are able to define what success and failure means before the dice are rolled. If you create too much granularity, it becomes difficult to decide what the result actually means.

True.  If you're left scratching your head after the dice are rolled, you have a problem.  

Quote from: Dr_AvalancheOf course, granularity might only be needed sometimes - in combat, for example. It could decide both who wins a fight, who achieved their objectives, and how injured they are.

Yeah -- combat is the biggest reason I was thinking of granularity.

Quote from: Dr_AvalancheI'm also wondering if this would benefit from open-ended dice, so if we're assuming d6's, each 6 would give both a success and a re-roll. That way, small dice pools wouldn't feel quite so discouraging. Not that the dice pools have to be small, but it seems more elegant somehow than rolling 20 dice for no particularly good reason.

Hmm.  Yeah, I think you have a point there.  I'm used to rolling buckets of dice, but if you've got two dice pools, things could get even more obnoxious really quickly.

Okay, so you have a basic dice mechanic and a general idea of how you want the rules to look.  Where do we go from here?

Quote from: MaddmanOne issue I see right away is that the goal pool is far, far more important than the skill pool.  After all, if the goal pool fails then you fail even if the skill roll succeeds.  And if the goal pool succeeds then then you succeed, even if the skill roll fails.  You may consider this a feature and not a bug but it leads to the obvious question - why have skills other than flavor?  If you can make sure you have a good goal pool then it doesn't matter what skills your character has.

I'll get back to the rest of the post later if I can add something constructive. The above is important. I don't think it's a feature that the skill is "just" flavor. It might do something else, which I'm not really sure of right now... Maybe success in the skill roll:

- Adds one success to the Goal roll, ensuring it's not "botched" (a concept I'm not sure if it should be in the game, but still...)
- Gives XP
- Limits just how bad Goal failure can be
- Something completely different

Quote from: GunhildaOkay, so you have a basic dice mechanic and a general idea of how you want the rules to look. Where do we go from here?

If I knew.

Is this interesting enough to actually create the system around it that it would require? I must admit that while I didn't have that in mind at all when I wrote the original post, it's a little tempting now.

Design-wise, I'm not sure where to go next. Maybe take a step back and try to put the system in a context. Does it need a "setting", even if that setting is as generic as saying "this is a fantasy roleplaying game. It's not built for science-fiction"? Maybe look into how people interact, when dice pool meets dice pool.

There's so much that would need to be done if I decided to develop this into an actual game that I literally don't know where to start.

Well, I can start by thanking you guys for bouncing ideas with me. It's really helpful, even when it's "well, that's cool, but not what I had in mind".