I know there are people on here whose math is way better than mine

[bbblakey]

and I'm about to prove it. My 10+ year old high-school maths has abandoned me (in fact, I think we did this stuff when I was 16. Yipe!) and I think all my textbooks are long gone, too

Comments

[truebluespark]

Hmm, well, it's been ages since I had a proper math class myself, but I can at least suggest it might be easier if you looked at it from the opposite direction. That is to say, find the probability that all of them will return false (in other words, the probability that zero of them will return true), then subtract that probability from 1.

Let's take the coin-flipping example. The odds that one coin will come up tails is 1/2, so the odds of all of them coming up tails at the same time is 1/2 x 1/2 x 1/2 x 1/2 = 1/16. That means the chance that at least one will come up heads is 1 - 1/16 = 15/16.

Hopefully that helps at least a little! I'm kinda tired and headachey at the moment, so that's about as in-depth as I can manage. And if you're looking for the probability that only one will return true, instead of at least one, that'd probably be out of my range of expertise.

[bbblakey]

That does ring a bell, thanks!

[tyoko]

This is probably way too late, but what you need is the Binomial Distribution:
StatTrek
Excel

For your coin example, the excel formula would be '=binomdist(1, 4, 0.5, false)' which corresponds to the probability that 1 coin out of 4 with turn up heads, each coin having 0.5 probability. The false on the end turns off the cumulative property, so we get the probability for getting exactly one heads-up, rather than up to one heads-up.

You may also want to look at the Poisson Distribution:
StatTrek
Excel

Edit: fixed some borked links

[thalass]

Planning to build an Infinite Probability Drive, are you?