@marty some people are consistently better than others at scissors/paper/stone. It isn't just random, though either side can force it to be random by making their choice randomly. However, it is a lot more difficult to make a random choice than most people realise. The temptation is there, even if you roll a 1 and that should be scissors, to roll again because you think your opponent is probably going to go stone. Most people succumb to that temptation far more often than they are willing to admit.
@Yakman Communication is there just from the hunt results. You know you went odd and the result was even because you missed out on the hunt. It is a lot weaker than a sorc trying to negotiate as a middleman between two packs, but it is there.
@kriszo The matrix can be generalised to produce any probability by choosing the right modulus and payoffs. As long as it is a zero sum game (as in the examples we have been discussing so far) it works. You can't have positive sum games (like euro's chance of a double hunt) because the participants could cooperate to rig it against the village. A chance of a double hunt would have to have a villager input in some way and hence be turned into a zero sum game but I haven't thought of a good way to do that.
@sbr Things like the SA scanner interactions can also be arranged via number games. The hunter defense is a simple extension to the odd/even game. The hunter's number interacts with the wolfs number, and the wolf goes down or not. The SA's number interacts with the scanner's number and whatever information that dictates is supplied. I think there are good chaotic opportunites for the SA being a number game. A player says in the thread, "I am SA and my number is x". The scanner can then choose a number to give away their identity or not, except that the SA player might be lying ... There is a point at which it gets too complicated for the non-mathematical player, but the GM just needs to tell the player what the appropriate modulus is for their role/traits and they can then force the result to be random by choosing a random number in that range. However, if most players are doing that most of the time, it rather defeats the point, better one number game they engage with, than several that make it too confusing to engage with any of them, so practically it will make sense for the GM to roll the dice on some things. Just as long as GMs get the message that the first time they roll the dice, they are a BAD GM, the second time they are a BAD, BAD GM and so on.
@OY percentage chance to fail for a scanner is tricky, but can be done if you can decide who the right participants are. However, I turn it back on you and say, why should there be a percentage chance? If it is really important for the game dynamic to have a percentage chance, then it will be clear who the opponents should be, and the appropriate number game can be set up between them. If it is not important then cut it out of the set up. As an example, let us say you want a game in which the seer will have a 25% chance of mistaking a wolf for a villager. Wolfs and the seer must give a number to the GM, and if they sum to 0,1 or 2 mod 4, the wolf is detected, and if they sum to 3 mod 4 the wolf looks like a villager. Let us suppose the wolves all go with 1. There comes a day when the JL says "lynch the scanned wolf". The remaining wolves change their number, ... or do they ... perhaps the seer will expect them to change and will also change his number so they should stay the same ... Thus you have chaos rather than randomness in the game. If this failure chance on the scans is worth having in the game, then it is worth the effort to make it chaotic. If it isn't worth the trouble making it chaotic, cut it out of the set up and make the scan an all or nothing result.
o
