As much as I like Math, Probabillity has always been one of the subjects that I struggle to develop an intuitive sense for. It's really something I should sit down and relearn one of these days...
Anyways, I have a puzzle to present to anyone who would like to try to solve it. Its solution, or any ideas that could help me reach the solution, would be incredibly helpful for developing an Extra Credit Artificial Intelligence that can play a particular Card Game.
Here is the problem:
1) You have a standard card deck with 52 cards, 4 suits, cards 2 through Ace per suit, standard stuff.
2) You are dealt a hand of size N, that contains A 4 of a Kinds, B 3 of a Kinds, C 2 of a Kinds, and D singles. So 4*A + 3*B + 2*C + D = N.
3) Given the hand size N, what is the probability of drawing a particular combination of A, B, C, and D? Assume that your input is always correct (so 4*A + 3*B + 2*C + D = N).
That might be phrased awkwardly, so feel free to ask questions if that doesn't make any sense. I couldn't find any easy to use resources to solve it on my own in about 20 minutes of research, and really, anything, even a pointer in the right direction, will help tremendously.
Thanks!
-
Apologies for necro - I was pointed at these boards today by a friend from another fora, thought that anyone with similar probability question might find an actual answer useful. This can be done via a trivial combination of the hypergeomentric distribution and permutation counts. The formula for the exact probability of a specific outcome given the conditions of the OP then simplifies to (sorry for so many parenthesis/etc. - no gaming fora I've seen supports mathematical markup properly): (25025*(3^(c+5))*(2^(2*b+c+2*d+10)))/ (a!*b!*c!*d!*(13-a-b-c-d)!*Bin(52,4*a+3*b+2*c+d)) Where a,b,c, and d correspond to the OP definitions, and Bin(x,y) is the binomial coefficient (x choose y). Improper values (e.g. asking for 14 singlets) taken care of "automagically" - you end up with an undefined mathematical operation.