Would you pay for a poker-playing robot program that may help you win in Texas Hold’em?
Texas Hold’em is a variation of a game of poker where there are four betting rounds: two cards are dealt face down to each player and five cards, “community cards” are dealt face up, the first three community cards are dealt in what is called “the flop”, an additional card is dealt in what is called “the turn”, and the last card is “the river”. A player, during his turn, can choose to raise, fold, or call any bets at each round.
Michael Bowling, a professor of computer science at the University of Alberta in Edmonton, developed a poker-playing robot program that gives a player his best response in a game of Texas Hold’em. Although this is a significant breakthrough marking “the largest game of its kind to be solved by an algorithm” (there are 319 trillion possible moves), the research concluded that the robot can be defeated in an actual game and that the first player or the dealer has a greater advantage because he also makes the last move. The reason is, unlike a game of chess where there is perfect information, players do not have all the information they need; you never know what cards your opponents are holding until you come to that final faceoff. You can memorize your opponents betting history and rely on that information to make your next move, and pray that luck is on your side.
It boils down to this: like most card games, luck is a factor in your chance of winning. There can even be 8 to 9 players in a game of Hold’em, which makes it difficult to memorize opponents betting history. Your ability to bluff also matters. Regardless, I find it remarkable that a program was developed with the best response for a game with imperfect information. I was not able to retrieve the results of the study from Science because I am not a member, but to all avid players of Hold’em, your best bet is to check out the findings because it may help you win the pot the next time you play.
(Disclaimer: Please note that the game Bowling weakly solved for is only one variation of Hold’em; other Hold’em games have different rules.)
http://www.philly.com/philly/news/science/Programmers_take_on_Texas_Hold_em_and_Win.html
Economic Theory of Networks at Temple University 
Wow, an algorithm that can process 319 trillion moves with imperfect information. Although Poker is somewhat trivial, I think this algorithm demonstrates how game theory can solve complex problems.
It was also interesting how the article quoted Alan Turing, who is featured in the movie that just came out The Imitation Game (Benedict Cumberbatch plays his role). It is a reminder of how most innovations happen at the crossroads of economics, mathematics and computer science.
First off, it is amazing to see just how applicable what we are learning really is. This being said the program that can play Texas Hold’em using game theory is truly impressive. Programs that just play the game of Texas Hold’em have been around for quite awhile. Many of these are just programs that run in virtual games against a human opponent. For the most part nothing really fancy occurs other than the computer running probabilities, and playing or folding depending on the probability of its hand. The big issue with this is that it leaves out a lot of available information that comes from the other players. This is exactly what makes this program special, it uses game theory to play more like a human and thus increases its chances of winning. It is astonishing that a program can account not only for information within the game, but also information from other players playing the game. Now in respect to the question of whether or not to pay for such a program in order help one’s own results. I think understanding how the program makes its decisions would be beneficial, but the program itself cannot be used to win for you. This is unless you are a casino and purchase this program to run as the computer player in their virtual games. A program like this one would allow the casinos to win more hands as compared to their old system. So, overall it is impressive that this program was created, and it is going to be interesting to see how it is actually used. What I cannot wait to see is what game will be solved next using game theory. This is because the games in which they are applying and successfully solving using game theory are becoming increasingly more difficult.