![]() The problem of getting data on all opponents (10 billion hands in your HUD) is overcome by software that was once legal. There is software that can select the chosen ranges your opponents statistically hold, live, at the tables. At first glance it doesn't seem to make sense that 99 has more equity than JJ pre-flop. Note the odds in the bottom 3 charts for 99, TT & JJ. I've made an example that shows the odds of a particular hand winning vs a range of hands. I like ProPokerTools because it calculates equities against ranges and exact hands for the games I play. Many good simulation tools are available. if the given board is like Ah Kd 7c, 2 more cards are added randomly for every simulation to produce always a complete 7-card sample to define the winner in the showdown that is the winning chance, it's always defined with all cards shown)įor each simulation, you have a winner between 2+ hands these results are stored internally, then they're divided with number of simulations to produce the so called winning chance. The third step is for every missing board card (etc. The same happens for every other player (if they're filled). For example, if for a player the input range is like 22+,A2o+, then it breaks down the whole range into something like 22,33,44,55,66.A2o,A3o,A4o., then picks a random hand from it, that become the hand for the current simulation. A specific board (up-right of pokerstove)įor each simulation, pokerstove takes a random hand from the input per player. ![]() A specific hand(s) (empty ranges of pokerstove in the left).The second thing is then to run a great number of simulations (in the magnitude of tens or hundreds thousands or even millions), having these inputs: That needs also to be fast, like something CactusKev evaluator. The first thing they do is to create / use a library that allows them to compare a 5-6-7 card sample vs another 5-6-7 card sample to calculate who's the winner of the 2 hands. I don't have inside information about the mechanisms Pokerstove uses to calculate the winning chances for each range but it's rather straightforward to compute them.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |