Talk:Machines

I just encountered a reroll machine in a regular room. It does nothing but it is.
 * I'm assuming that if there are pickups or miraculously items around it'll have effect. --SirMoogle (talk) 01:40, 29 January 2017 (UTC)

Does anyone have any information to add about the blood donation machine's chances of breaking and whether there are conditions where it cannot break? I have tried to break one now well over 100 times with The Keeper and wondering if this is atrocious luck or if it's not supposed to break with him. 208.98.223.30 06:44, 23 December 2019 (UTC)
 * Pretty sure it's just down to luck, haven't found anything that would stop them from blowing up. Frionil (talk) 19:22, 23 December 2019 (UTC)

It is reported that the crane game pays out 25% of the times, however if this is true then the expected number of plays for each reward should be 4, not 2.41 as reported (since the probability of winning after exactly n tries would be (3/4)^(n-1)*(1/4), and so the expected number of tries should be the sum of n*(3/4)^(n-1)*(1/4) over n positive integer, which is 4); the formula reported in the article (that is, 1-(3/4)^N) would be the chance of getting the reward after at most N tries. Am I missing something here or am I correct in believing the article, in its current form, is incorrect? --Biancospino (talk) 20:45, 12 May 2021 (UTC)
 * I think the problem is with the phrase "Average number of plays for a single payout is 2.41". 2.41 is the exact number of plays you'd have to do to have a 50% chance of a payout, but isn't what most people would think 'average plays needed for a payout' means. Even 4 plays only net you a 68.4% chance of a prize. Maybe a short table of percentages would work better? --108.66.200.211 22:47, 18 May 2021 (UTC)
 * Ok, but by all mean it is not a meaningful quantity. I mean, the number of e.g. coin tosses necessary for 50% chance of a tail is 1, but the average number of tosses per tail is, by any reasonable definition, 2. Also, I agree with the table idea. I'm not entirely sure if it would be better to have it in the form (n | probablity of winning in at most n tries) or (n | probability of winning in exactly n tries), but the former would seem nicer to me; certainly the current formulation is at best dubious and at worst plain wrong. For now I've removed the 2.41 claim. --Biancospino (talk) 21:49, 19 May 2021 (UTC)