Subj : Re: cards To : comp.programming From : mschaef Date : Wed Oct 05 2005 09:24 am In article , Omri Barel wrote: ... >In the loops solution (seven nested loops), the last variable never >takes the first value. Translating to cards, it means that you're >assuming that the last card on the table can never be the ace of spades. > >This assumption would change the probabilities considerably (and is >definitely wrong for a game of poker). > >So if you want a call/fold/raise winning probabilities after each card, >you can't just go through all possible 7-card sets, you have to look at >the order of the cards in each set as well. I'm not sure it's quite that bad. Each player starts out with 2 private cards, and then five public cards are made available in three groups: the flop (the first 3 cards), the turn (the fourth card), and the river (the last card). So, the set of 7 cards is made of 4 subsets: Size Name ===== ===== 2 Private (This is unique to each player) 3 Flop 1 Turn 1 River Within the Private and Flop sets, ordering doesn't matter. This reduces the search space by a factor of 12. One complication in all this is that the value of a given set of 7 cards is strongly influenced by the potential value of the public cards by themselves. That is, if I'm holding a pair of 2's, they aren't worth as much if the public cards look like 4,4,3,3,3 or 4,5,6,7,10. I would expect considerations like those to have a big role in deciding the overall worth of a hand. -Mike -- http://www.mschaef.com .