Step-die pool mechanic for consideration

[Fallen_Icarus]

Hi all.  I've been away for quite a while so bear with me if I'm not up to date on all the recent threads.  

The core mechanic for the game I'm designing centers around a pool of dice that represent each character's potential ability.  The dice vary in size.  For example, one character might have a pool that consists of two d20, four d10, and one d4, while another (and much more formidable) character might have one d20, one d8, three d6, and two d4.

As you might have guessed, the idea is to roll low.  Target numbers are determined by adding the appropriate Stat value to the Skill value.  A player trying to research something, for instance, would add her Mind Stat (3) to her Research Skill (4) and come up with a Target # (7).

For difficult tasks the GM may reduce the Target # or may at times roll a die and add it to the players roll.

A player may choose to roll any of the dice in the pool, but if the highest value is rolled on that die it is spent and thereafter, unavalible.  Dice can be regained but not without further risk.

Would anyone mind helping me with finding probabilitys for this system?  Being that it uses multiple die sizes and has fluctuating target numbers I have a hard time knowing where to start.

Also, how does the mechanic rub you in general?  Are there any obvious problems that you can see cropping up during play?

Do any other games use a similar system?

Thanks for the help,
Eric

[Jack Spencer Jr]

The problem with a dice-step pool vs a target number, like what you've got here, is that it's difficult to balance the target number correctly against all of the different types of dice. Using you example target numer (7), d6 and d4 would be an automatic success.

(Personally, for a multi-dice system I would prefer something like Button Men, but that game is copyrighted.

[failrate]

Of course those would be automatic successes, but each has a higher probability of the die rolling high and being lost.  The only thing I'd do different if I were designing the same kind of system would be to reverse the high/low thing, but that's mainly because I find it more gratifying to roll really high numbers.  Psychologically, the higher number *feels* more like a winner to me.  The mechanics would be relatively unaltered (except there would no longer be automatic successes).

[Andrew Martin]

As your system seems to be a "generic system", what reasons are there to use your system over, say, Fuzion?

[Walt Freitag]

Would anyone mind helping me with finding probabilitys for this system? Being that it uses multiple die sizes and has fluctuating target numbers I have a hard time knowing where to start.

Let's start by clarifying the rules here. Especially, you have to tell me what is not variable.

In both your examples the players have a total of seven dice in their pool. Do players' pools always have seven dice when fully replenished (that is, no dice currently "lost")?

Does any action succeed if any single die is rolled below the TN? Or do some actions require more than one low die to succeed?

For any given dice pool and target number, the probability of getting at least one success is pretty easy to calculate, and the probability of getting some specific number of successes isn't that much harder. The problem is that there are 330 distinct ways of making up a 7 die dice pool using five different possible die sizes. Far more, if d12s are included making six different possible die sizes. That makes it kind of hard to tabulate the probabilities in any convenient or enlightening way. Is there any other constraint on the dice pool composition (such as a minimum or maximum total number of sides) that might reduce this to a more manageable set of cases?

- Walt

[Fallen_Icarus]

The Stat values and the Skill values will usually stay pretty static, advancment and modifiers aside.  However those numbers are assigned at character creation and are not standard.  Max value on both Stats and Skills is 10. Average human being a 5.

The number of dice in a pool is also not set.  I would like to see most players start play with a pool of about 5-7 dice.

All actions are resolved using a single die roll.  As pointed out if the max die roll is lower than the target, success is automatic, but if the max value is in fact rolled the die is lost.  Characters who give it there all will get burned out quick.

There is no (as of yet) side limit to the dice.  All dice are allowed.  I know this makes it hard to tabulate but the mechanic itself is fairly straightforward.

I'm not familiar with Fuzion.  This game is not going to be "generic" in the end, though.  I just need to hash out the system before I add the specifics.

I like rolling high as well, failrate, but I've found that a roll-under method keeps the system a lot tighter and more managable.

EVH

[Valamir]

I'm a fan of clever dice mechanics, and this one seems pretty clever.
Here are some issues:

1) The principle draw of the system is in modeling depleting effectiveness.  As time goes on the character gets less and less effective.  Presumeably the goal is that through skillful management of the die pool early on you wind up with enough low dice left at the end to defeat the final obstacle.

I would encourage you to make sure that the game world you use this for is one where that is an effective model.  You rightly point out that this is not a generic system.  It is in fact a pretty narrow system.  It would not be good at all for modeling high impact hollywood action style adventures.  In those the character effectiveness generally escalates through the story peaking in a climactic battle, best mechanically illustrated by Extreme Vengeance.

The stories that should be told with this system should be stories of weariness of hardship and struggling to get through.  I could see a system like this used to play a platoon of soldiers heading in country in Vietnam where by the time they get back they are down to their last couple of dice.  

2) The system could easily be (and perhaps would best be) suited for some serious abstraction.  By this I mean, the entire platoon of soldiers in nam could be represented by a single die pool, with each soldier contributing a die of differing size, and each lost die translating directly to a casualty.  Rely on the heroic Sarge to save the day too often and Sarge dies.  Try to make do with the green replacements and roll a lot of crap.

3) The system pretty much requires alot of rolls to come into its own.  The odds of losing effectiveness by losing both of your d4s is not that worrisome if you only make 3 rolls the entire night.  Thus die rolls that encapsulate a lot of action into a single roll might not be the best option as this leads generally to fewer rolls.  I'm thinking the resolution almost requires being task based rather than conflict based.


As an alternative way of approaching the idea:
1) Consider rating every trait or skill with a die size and then assembling your die pool in Iron Claw fashion by simply selecting every trait on the sheet that might be influencial for a particular roll and rolling it.
2) When a die rolls its maximum value its not lost, rather that trait or skill loses a level of effectiveness...bounces to the next higher die type, representing exhaustion setting in.

So a medic trying to heal a patient might grab his d8 intellect, his d10 medic training, his d12 medic equipment, his d6 Will, and then...because this patient is important to him, throw in his d4 Luck also.

He makes his roll and his will die came up a 6, while his medic equipment die came up a 12.  His Will drops to a d8 (explained as mental fatigue) and his medic equipment (now largely used up) drops to a d20.

[Mike Holmes]

There's something here that very much needs to be clarified.

By "one die roll" do you mean "the roll of one die" or "one roll of as many dice as you like"? If the first is the case, then the probabilities are quite simple. If the latter is true then there's another problem. Which is that you say that you take the "max" value for the roll. This sounds like you mean the highest. But since it's a rollunder system, that would make it non-sensical to roll more than one die ever.

So is it a one die per check system, or do you take the minimum value to determine success?

Mike

[gabby2600]

Humm...


Sounds a simialr concept to the lastest Incarnationof Marvel, Their card system works quite well and when I played I really enjoyed the game.

The dice thing works but I think is a little to depressing as you watch your pool plumt and if you know your going to role-play bad then your have no hope in your system.

How dose this knock you.

Big black bag-o-numbers, each player picks 5 numbers (random number 1 - 20)  once you have used a number counter you discard it (back to the bag) and pick another random number.  This is less frustrateing and random, also a lot faster, because you still have that chance to fail and I think saveing your best dise tioll last and failing vs succeding on your worst is a pain.

Also with the number counter method you an taylor the numbers to suit, so you cna have a few v.low and v.high scores and lots of mid range scores. Or even the choice of numbers for the bag could be determinde when the game starts.

This give the payers a choice to know they will succed at something or they ca let the chekc slip and fail it. for a hero game this works really well. it might work for your game.

[TheRedSoup]

I think Mike Holmes is correct.  What would be the MAX roll?

It seems that if you ever roll a ONE then you lose the die.  You can call it "The Rule of One" or some such.  Easy to remember, I think.

[Fallen_Icarus]

Sorry for the confusion.

Only one die is ever rolled at a time (at least with the system as it stands now).  The object is to roll low, under a target number created by adding Stat and Skill.  If that die ever rolls its highest value, however (4 for a d4, 6 for a d6...) that die is lost, AKA booted out of the pool.  This happens even if the roll succeeded.

Eric

[Mike Holmes]

Probability is then Target Number divided by sides of dice rolled. Thus if my TN is 3, and I'm rolling a d6, my chances for success are 3/6=50%

Was there a more complicated problem that I'm missing?

Mike

[Valamir]

ok

[Fallen_Icarus]

Valamir, thanks for the feedback.  That gives me a lot to think about.

The stories that should be told with this system should be stories of weariness of hardship and struggling to get through. I could see a system like this used to play a platoon of soldiers heading in country in Vietnam where by the time they get back they are down to their last couple of dice.

Absolutley.  I want this game to be about struggle in the face of adversity and heroics against seemingly insurmountable odds.

The system could easily be (and perhaps would best be) suited for some serious abstraction. By this I mean, the entire platoon of soldiers in nam could be represented by a single die pool, with each soldier contributing a die of differing size, and each lost die translating directly to a casualty.

I hadn't thought about this aspect of the system before.  It could easily be used for mass combat, or even ship to ship combat where the dice represent gun turrets and engine rooms.

Thus die rolls that encapsulate a lot of action into a single roll might not be the best option as this leads generally to fewer rolls. I'm thinking the resolution almost requires being task based rather than conflict based.

This was my thought exactly.  Action needs to be per task in order for there to be a risk involved.  I also am curious to see how it holds up in combat.  Each roll depicting a blow, and damage actually reduceing an opponants dice pool.

Before I forget, I do have a preliminary means of regaining dice through the course of the game.  basicly, when a player runs out of dice he gets his pool back right away but has to remove his smallest (or best) die.  This spiraling method ensures that players will have to, at some point, roll all their dice.  It also supplies a means of simulating exhaustion.  Combat damage on the other hand reduces a players pool indefinitly.


Eric

[Walt Freitag]

Yeah, like Mike said, if it's one die at a time and it's the player's free choice of any die from the available pool, the probabilities are real simple. To lay them all out as percentages:

Code Select Expand

IF YOU HAVE TO ROLL THE TARGET NUMBER OR UNDER:

TN         d4         d6         d8         d10        d12        d20

1         0.25%      0.17%      0.12%      0.10%      0.08%      0.05%
2         0.50%      0.33%      0.25%      0.20%      0.17%      0.10%
3         0.75%      0.50%      0.38%      0.30%      0.25%      0.15%
4       certain      0.67%      0.50%      0.40%      0.33%      0.20%
5       certain      0.83%      0.62%      0.50%      0.42%      0.25%
6       certain    certain      0.75%      0.60%      0.50%      0.30%
7       certain    certain      0.88%      0.70%      0.58%      0.35%
8       certain    certain    certain      0.80%      0.67%      0.40%
9       certain    certain    certain      0.90%      0.75%      0.45%
10      certain    certain    certain    certain      0.83%      0.50%
11      certain    certain    certain    certain      0.92%      0.55%
12      certain    certain    certain    certain    certain      0.60%
13      certain    certain    certain    certain    certain      0.65%
14      certain    certain    certain    certain    certain      0.70%
15      certain    certain    certain    certain    certain      0.75%
16      certain    certain    certain    certain    certain      0.80%
17      certain    certain    certain    certain    certain      0.85%
18      certain    certain    certain    certain    certain      0.90%
19      certain    certain    certain    certain    certain      0.95%
20      certain    certain    certain    certain    certain    certain


IF YOU HAVE TO ROLL STRICTLY UNDER (NOT EQUAL) THE TARGET NUMBER:

TN         d4         d6         d8         d10        d12        d20

1          zero       zero       zero       zero       zero       zero
2         0.25%      0.17%      0.12%      0.10%      0.08%      0.05%
3         0.50%      0.33%      0.25%      0.20%      0.17%      0.10%
4         0.75%      0.50%      0.38%      0.30%      0.25%      0.15%
5       certain      0.67%      0.50%      0.40%      0.33%      0.20%
6       certain      0.83%      0.62%      0.50%      0.42%      0.25%
7       certain    certain      0.75%      0.60%      0.50%      0.30%
8       certain    certain      0.88%      0.70%      0.58%      0.35%
9       certain    certain    certain      0.80%      0.67%      0.40%
10      certain    certain    certain      0.90%      0.75%      0.45%
11      certain    certain    certain    certain      0.83%      0.50%
12      certain    certain    certain    certain      0.92%      0.55%
13      certain    certain    certain    certain    certain      0.60%
14      certain    certain    certain    certain    certain      0.65%
15      certain    certain    certain    certain    certain      0.70%
16      certain    certain    certain    certain    certain      0.75%
17      certain    certain    certain    certain    certain      0.80%
18      certain    certain    certain    certain    certain      0.85%
19      certain    certain    certain    certain    certain      0.90%
20      certain    certain    certain    certain    certain      0.95%

- Walt