CodePlexProject Hosting for Open Source Software

You are dealt two cards of 52 cards from the deck, opponents also have two cards. But you don't know which cards they have, which cards will be dealt to the table. Application deals random cards from the rest to your opponents, and 5 cards to the table. After that it compares each player hand, and decides who wins. After that montepoker plays another game with the same your cards, dealing other cards to your opponents and table.

If cards have been dealt to the table you can also specify it, and montepoker will play games with these cards.

This approach allows to get as much statistics information as you may want, where direct analsys may make you head blow. For instance, if you got two clubs, for is the probability that you will get flush? That's realy easy: You have two, eleven are left, so you can make couple of divisions, summs and get this. But what is the probability that your opponent doesn't have higher flush? That's become more complicated. And you also may want to need to know if I got flush what is the probability that at least one of the opponents has full house or four of kind, and so on. You can spend several hours of solving that and tons of paper, and .... make couple of mistakes in the middle..

Monte Carlo method is an approximatation, but give it just second or two and you'll get required statistics with less than 1% error. That's more than enough

Last edited Dec 10, 2012 at 8:25 AM by SergeySavchuk, version 2