Kleene Onigiri wrote:No, they are conditional.
Because if the spy knows that for example, team 1 get's voted most likely, when someone just mentions his random vote, he won't send a kill for that team (or at least, the chance that he'll try to kill them will sink dramatically) So the vote on a random team is influencing the chance of killing a team.
Also, as I said, it's not totally random :x Considering the spy didn't even kill at all, that chance that this would happen wasn't calculated at all in our calculations D:
These calculations were made on the basis of a random vote and I mentioned as much in my original post. If everyone votes randomly without telling us who they'll vote for, there's no way of knowing who will get arrested and in that case the two are completely independent. I'll agree that this will change if people say who they'll vote for, making it less likely for the spy to kill those people that are mentioned. But here's the thing: Giving the spy information to let him avoid killing the person who'll be arrested will decrease P(A and B), i.e. the probability of the detective getting arrested and being the kill target. Reducing P(A and B) however will increase P(A or B) = P(A) + P(B) - P(A and B), so what you're actually saying here is that my estimation for the detective being arrested or killed was too low.
Regarding the non-randomness of the vote: You say it's not completely random because people vote from their lists, but the lists, too, are random, with a slightly higher chance for the both the detective and the spy in comparison with the other players. Again, this was a fact which I mentioned in my original post as something which I didn't include in my calculation, but just like the revealing of votes, this actually increases P(A), the chance of the detective being arrested, because he has a higher chance of being on any given list, thus he has a higher chance of being voted.
Also, I'll admit that I didn't include the chance of the spy not killing, since I expected him to kill, so just imagine the following preface to my calculations:
Given a random vote and the spy actually killing a person
Anyway, I don't mean to say the calculations were perfect. I already admitted yesterday that I forgot the possibility of a tie occurring naturally, for example. Nonetheless, I'm sure that they're pretty close to the truth, and most of the adjustments I made (neglecting announced votes and increased chances of the detective and spy being on the list) actually kept my calculations on the conservative side, instead of trying to embellish the results and erring on the side of making the probabilities too high.
In any case, maybe we should try to return to the actual game now and think about our next steps. XD