After a session of Sorcerer a player and I got into a casual discussion of dice probabilities and system.

His contention was that in the recent White Wolf WoD games, with stable target numbers and with the magic of exploding 10's factored in, committing 3 dice to a roll is equivalent to 1 certain* success (* "certain" = 18.5 times out of 20, let's say).

Has this equivalence (3 dice = 1 certain* success) been observed empirically in your WoD play?

Has this equivalence affected practical gameplay decisions?

Has anything like this been raised to a consciously-manipulated currency? (In practice, mind you: the word "currency" might never have come up but currency-inflicted behaviours might have taken place).

If this is the case I have to revisit my gameplay with WoD.  I got caught up in the minutiae of all the powerz and the details of the factions, and got lost.  But seeing a functioning virtual currency at work int them game might allow me to step up my game on those occasions when I step into the WoD.

Since you mentioned Exalted, I would add that in a game I played several years ago, my Abyssal (eclipse caste?) had a Devil of a time getting anything done at all, when it came to rolling dice to do it.
I don't know if my build was suboptimal or what, but one thing that seemed to hamstring the dice system was a lack (iirc) of some resource I could spend to affect the outcome of the roll, after the roll had been completed. Willpower refills were hard to come by in this particular campaign, and I may have been tapped out pretty quickly.
My memory is regrettably hazy on the subject; forgive me if there's a hole in this argument. Also, lack of sleep.

Thanks. 

Your case is the same as mine.

I make these builds and I am not sure if I am really utilizing the mechanics.

I build these funnycars with chrome and extra lasers and air-brush art but I can't seem to pop the hood and jimmy with the engine.

I think part of the issue is that the system's priorities are either muddled or unclear.
If Exalted were laid out more like 4th ed. D&D, and we were explicitly told, "Hey, so design a character that's badass, so that he can kick asses in GM-provided scenarios," then I'd have been thinking about character optimization.
But Exalted's "how to play" text (not the rules, but what you do with them) is kind of a mess - like a lot of games out there, it claims to be part storytelling, part game, part Imagination Station, all things to all people. That's a product of its time period more than any specific fault of the writers, imo.

Anyway - something tells me that if you approach Exalted from a "power-gamer" perspective, i.e. trying to maximize resources, basically treating character creation like a puzzle to solve with the greatest amount of payoff you can get, it'd probably work just fine. I would consider it an element of Stepping On Up to see how well you can pop said hood and jimmy with the aforementioned engine.
Just so I know, is that what you're after? Based on this line from the "Why Dungeons?" thread,

Erik said,
I used to think that I was interested in stories that D&D couldn't tell but what I was really frustrated with was people who were messing with the parameters that made a game gamey...

If it is like an auto success, then it kind of ends up like Amber, from what I've heard? Ie, compare static value against static value. Though I don't understand that - to me it just seems alot of waving numbers around to obscure how the GM decides utterly (as he decides the number your comparing against).

Hey Callan! ^__^
The Feats of Strength subsystem might be useful to mention, here - at least the one used in Vampire: the Masquerade. If you have a Strength of rating X, you can automatically perform all Feats of Strength rated at X or lower (or maybe X-1 or lower?).
Thus, it could be that you can opt not to roll against a sufficiently weaker opponent, with Willpower points spent in a kind of bidding war. But I'm not sure I get the utility of taking the system in that direction. Just thinkin' out loud!

As for Amber itself, I think you're right about it devolving to GM-fiat; that's my relatively uninformed opinion, though. That's an issue I witnessed in the diceless RPG Nobilis, too - there's no explicit group-consensus mechanic to determine how powerful adversaries are or can be, but I'm not sure what to do about this issue. Since the CA best supported by WoD seems pretty clearly to be Right to Dream, I think the GM should be encouraged to make the case that a given adversary's capabilities make sense entirely within the context of genre/setting expectations.

An Elder vamp should have really high Disciplines, but low Firearms and Drive, for instance. Hm.

Hullo, Abkajud!

"Makes sense to whom?" Is my usual song and dance at this point. People have very different ideas of what makes sense. I think the notion of 'right' in right to dream was to use a system to determine who it should make sense to. Universalis is an example of that, I think.

I might sound off topic, but in terms of raising to a consciously-manipulated currency, this is kind of the next step. First you notice that hey, you have these numbers where you'd have a consistant result, so you could use that as a whole system instead of the big squiggle of random events it seems! So there's this structure to grasp! Structure, at last! Except it's not there. The GM is so much at liberty to change things, it's as much a structure as a dictator of a country presents (whether he's a good dictator or not doesn't matter - one guy with all the power is one guy telling all the story).

Though I was idling around the idea of a system - it was for a browser game though. The idea is that in a poll format a situation and several actions are described and players can vote for just one. Perhaps actions contributed by players. Okay, polling happens then after, perhaps quite some time after, the player might hit that situation. Okay, he can choose a move now, But he doesn't know which was voted the best move. So that taps into a group imagination, and is fixed, so no one has the capacity to decide which is the best move based on how they'd like things to go (ohh, no one ever does that of course - or - only evil GM's with no skillz do that!)

Okay, now I'm off topic, but atleast I described an idea that revolves around currency ;)

I'm reviewing the Exalted rules, and I'm a bit confused as to your statistics. For Exalted, 7,8,9 count as one success, and 0 counts as two successes, for an expected value of 0.5 for a single die roll. Regardless of this, the probability of success is somewhere less than 50-50. So for three dice rolled, the probability of at least one success is something less than 1 - 1/8, or 87.5%.

You say an "automatic success" for three die is something like 18.5/20, which is 92.5% (closer to a 4-die roll, 93.75%). Am I just being too literal on your math?

By the way, systems with "exploding zeros" would only increase these percentages slightly. (Math problem: can you make a system of exploding zeros where the expected value of the roll becomes undefined?) Sorry to hijack.

It's actually quite simply to calculate the odds of getting at least one success in a WW style dice pool, as that double success on tens does not affect whether you got any successes or not: assuming that 7-10 are successes on one die, we can conclude that you have a 60% chance of not getting that one success on one die. Therefore the probability of getting at least one success on three dice is 1-(0.6^3) = 78%.

Probability of getting exactly one success is also simple to calculate, incidentally: you only get exactly one success if no dice hit 10 and only one hits 7-9; the probability for a given die of three to hit 7-9 is 30%, the probability for the other two dice to miss is 0.6^2=0.49 and thus the probability of the entire sequence is 0.49*0.3=0.147. Because each of the three dice might be the one that hits, though, we triple that number to arrive at 44% as the probability of getting exactly one success out of three dice in the WW dice pool. Thus:
No successes - 22%
One success - 44%
More than one success - 34%

Regardless of that, I don't really follow Erik's logic regarding virtual currencies. Perhaps there is some point to it, but to me this seems like a basic property of randomized systems -they have certain probability distributions. There is no magical equivalence or guarantee of a success at three dice threshold if that's intended, though; adding dice increases your chances of succeeding, but doesn't remove the possibility of failure.

Regardless of that, I don't really follow Erik's logic regarding virtual currencies. Perhaps there is some point to it, but to me this seems like a basic property of randomized systems -they have certain probability distributions. There is no magical equivalence or guarantee of a success at three dice threshold if that's intended, though; adding dice increases your chances of succeeding, but doesn't remove the possibility of failure.

Actually, yes, I noticed that too.

Playing NWoD, I used to expect roughly one success per three dice. Playing Exalted, I used to expect roughly one success per two dice. In the latter, due to monstrous dice pools tending towards the average, this estimation was closer to the actual result more often.

My resource management strategy was strongly informed by those estimations. Also, I recall in both games we had house rules moving some dice boosting effects from before to after the roll. In Exalted, for instance, instead of declaring the use of defensive Charms before any dice hit the table, we used to only do that once the attacker's successes were known. I think we also used to spend Willpower after the roll (which wasn't that good an idea in hindsight, since it resulted in cancelling that single bomb-crafting botch that could have proved really interesting otherwise).

Now, I'm posting this just after a session of Mouse Guard and come to think about it, I've been doing that estimation thing a lot today. That is, when positive results were really needed, expecting roughly one success per two dice, I've been taking all steps to build a pool of more that twice the obstacle. When positive results weren't needed that much, and the pool was already at least twice the obstacle, I've been careful not to waste resources (or looking for opportunities for hindering the character to generate extra resources).

Incidentally, similarly to our Exalted and WoD house rules, in Mouse Guard you never start adding bonuses before the obstacle is set.

Erik wrote:

I got some real world feedback on an idea and that's all I was looking for.  Thanks y'all.

Fair enough, though that becomes the difference between statistics and psychology. For example, we all remember exceptional results much more easily than the mundane, the obvious. Perhaps that's why successes on two dice, in spite having a 1/3 chance of failure in Exalted, are seen by at least one poster as a bankable thing, since it seems so counterintuitive given that the odds are obviously against you with one die.

Possibly. But the same argument can be given the other way, I realize.

Here's another point, to defend my mathematical honor. Eero said correctly that the probability of failure for three dice (in Exalted) was 22%, simply cubing the probability of failure for one die (6/10). My statement was about 13%, obtained by cubing the expectation value of one die, which is 1/2. I did this because I believe expectation values to be more indicative of people's psychological interpretation of the dice roll than the direct answer to the question, "what is the probability of success/failure?". I was answering a slightly different question, "what is the probability of getting zero when rolling three dice, weighted by the amount of successes per roll?", which sounds rather silly, but I swear by it.

So if OP were to do this another way, one interesting question to ask the forum perusers is "When rolling two dice in Exalted, would you be surprised if you succeeded?"

And of course again I'm way off topic. Maybe I just have a problem with taking any risk of failure - probably why I don't play Poker.