At the online poker tables, you're pitted against other people in that age-old poker exercise of taking their money.
Sometimes, however, you have those nights when no bluff works, your second pair or ace-high bluff is picked off by another player's accurate play or you're run down on the river by a better hand.
While many players will just shut down the tables and write the night off, some ask the question, "Hey, am I playing against a bot?"
Rumors about the existence of pokerbots - computer-created programs that can play for hours on end without a break, that make the optimal move at all times and that allegedly can turn a profit - have permeated forums since the dawn of online poker.
Any Internet search engine will yield thousands of hits for products claiming to be this kind of program. The reality, however, is that the technology isn't quite up to speed when it comes to bots for full ring games and tournaments.
Computer whizzes have long been devising programs capable of challenging the best human gamers in the world. For games with few moves and deviations, such as checkers, Scrabble and Connect Four, these programs have been able to defeat even the best that the human race has to offer. For more complex games, it took a bit more development work.
Chess computers circa 1990 weren't able to even approach the grandmasters of the game. That all changed in 1997 when Deep Blue, a computer created by IBM expressly to play chess, defeated world chess champion Garry Kasparov in a hard-fought series of matches (although the computer's victory remains controversial to this day).
A poker program that can beat human players is one of the more recent targets of developers, and they've been making great strides toward that goal. The University of Alberta's Computer Poker Research Group has been at the forefront of this movement.
The U of A team first created Poki and PsOpti, aka SparBot (the latter of which has been used in both Daniel Negreanu's video game "Stacked" and the Poker Academy Pro computer program). Then, earlier this year, they unveiled Polaris. Of these programs, Polaris has shown the most promise (or threat, depending on how you look at it).
In July, Polaris squared off against champion poker players Phil Laak and Ali Eslami, with the criterion for victory being cash won or lost by each of the players. In a five-session series, the computer played both players individually heads-up, which yielded startling results.
Polaris was able to reach a statistical draw with the two poker pros in the first match before winning the second round. Laak and Eslami, though, were able to strike a blow for carbon-based life forms by winning the next three sessions, albeit by very small margins.
While they demonstrated that humans could still beat computers, the final score was close enough to indicate that programming had come a long way. That said, pokerbots playing online isn't something players should be excessively concerned about.
While there are several programs advertised online that state they can play just as well as the University of Alberta's, the reality is they cannot; it's merely marketing bafflegab. Even the creators of Polaris state that their program can achieve winning results heads-up only; it's at a loss confronted with multi-way hand or full table parameters.
Furthermore, they maintain that Polaris only works in a cash game situation and not in tournaments. (Admittedly, cash games are often more profitable than tournaments). Finally, the computers we are talking about here are full-fledged systems dedicated solely to the game at hand, requiring computing capacity in the terabytes to accurately deduce their strategy.
Online poker sites are also aware of the potential for pokerbot situations. The poker sites monitor length of time at the cash game tables and, upon coming across players who haven't moved from tables in an inhuman amount of time, disconnect them.
Pokerbots, though, are probably more of a scapegoat than a reality in online play. Cries of "pokerbot" may well be rooted in human beings' capacity for denial. Even beat after beat, it's hard to face the fact that you're just not playing at an optimal level.
Instead, players ignore how badly they're running and foist the blame on a program. The simplest explanation is discounted in favor of a more elaborate one that lets them off the hook. This isn't to say that, given inevitable advances in computing capability and artificial intelligence, functioning pokerbots are out of the question.
With an AI sophisticated enough to take into account other players' styles and adjust its own game, they could be viable. That technology, however, is likely still a few years away, making it futile for online players to worry about it at this point.
So the next time you feel you are up against a bot, take a long, hard look at your own play. Those miraculous 5% chances do come home on the river sometimes. There are players who can pick off your bluffs better than any machine could, and there are players capable of waging optimal poker war for long sessions. Pokerbots shouldn't be your primary concern.