On 21-Feb-2009, at 15:12, Jared ''Danger'' Earle wrote: > On 21 Feb 2009, at 21:14, LuKreme wrote: >> (yeah, I sorta skipped the statistics portion of my Mathematics >> Education) > > You must have never played pen-and-paper role-playing games, either. > If you can't figure out the chances of rolling 18 on 3d6, or working > out where the bell curve is, you need to get old school dice and get > learnin'.
No no, I can do THAT. I can figure out the Powerball odds even. The trouble is, when I don't know the mechanism, I don't know how to gauge the certainty of a particular number. Obviously 19/79 is a lot less certain than 1900/7900, but there must be a way to calculate the 'certainty' or margin of error or something. For example, if I do something 5 times successfully, I might think that the odds are 100%. But then if I fail the next 15 times, now the 'odds' have dropped to 25%. Let's say that the real number (that I can never know for sure) is 50%. As I keep doing it, that 100% dropped to 25% and will creep back up to near 50%. At some point it will effectively BE 50% and it will never move significantly, but at some point long before than one should be able to say, "The odds are 50%±3%." > Oh, with your example, get a piece of grid paper and plot the results. > If they're roughly a line, you know it's easy. If they're roughly a > curve, you know it's harder. If it's zig-zag, you know you're in > trouble. I only have two results, yes and no (or 'positive' and 'negative'), I'm not seeing how graphing it is going to be useful. I know 19/78, the order of those 19 can't possibly matter... -- I can't die, I haven't seen The Jolson Story - Jetboy _______________________________________________ OSX-Nutters mailing list | [email protected] http://lists.tit-wank.com/mailman/listinfo/osx-nutters List hosted at http://cat5.org/
