Kyle Siler Interview: Why Winning Isn't Everything in Poker
Online poker is one of the biggest internet-focused industries in the world. In 2004, DrKW and Global Betting and Gaming Consultants found that online poker revenues were worth just over $1 billion and since then, the industry has dramatically grown. The most recent edition of the iGaming Business Directory listed 235 websites which they found are the “top English language iGaming poker sites”, and as the list continues to grow, more and more players sign up in the hope of winning big. While few players manage to climb the heady heights of the poker world, Kyle Siler, a social scientist at Cornell University, has found that by aiming to win more hands, players often end up sacrificing more cash. In late 2009, Siler’s study on the “Social and Psychological Challenges of Poker” was published by the Journal of Gambling Studies and within this paper, the author discusses the aforementioned “counterintuitive incentive structure” which, on a basic level, involves winning hands and losing money. We spoke to Siler about his work, as well as the online poker industry and the effects his research may have on seasoned poker players and newcomers alike.
Siler chose to study poker as he “is interested in the contexts and contingencies that make different competitive strategies more successful and more popular” and due to the very nature of the card game, poker rooms made an apt environment to study such aforementioned features in detail. The “online poker boom” has meant that many more players are becoming involved in the game, as the “intimidation factor of facing card sharks and shady characters in back-rooms, or huffing down second, third, and eighth-hand smoke in a Las Vegas casino” has effectively been removed. Internet poker has also proved a more academically viable topic, as unlike its offline counterpart, online poker allows scientists such as Siler, using programs such as HandHQ.com and PokerTracker.com which record hand histories, to “study human behavior with real money and real decisions”. Siler chose to focus on No Limit Texas Hold’Em, as this poker variation “has a significant social-psychological component, in addition to the technicalities of the card game” and as yet, scientists have been unable to develop a bot which can beat the very human interactions which take place during the game. Using “27 million anonymized hand histories” from HandHQ.com, Siler soon found that “the larger percentage of hands one wins, the less money they tend to win”. While Siler’s study discusses in-depth the reasons behind this “counterintuitive incentive structure”, the scientist summarised them as follows:
“I theorized that the asymmetric incentive structure (frequent small wins, outweighed by occasional big losses) and not understanding the existence and logic of Pyrrhic victories (a victory which proves devastating for the victor), were what appeared to hurt losing players. Behavioral economics has shown that people have difficulty weighting occasional large losses vis-à-vis frequent small wins, in part because humans tend to perceive risks and gains differently. So, it appears that for whatever reason, people (or at least, losing poker players) tend not to evaluate their risks and rewards appropriately or proportionally.”
Online poker and the advent of HandHQ.com and PokerTracker.com software have meant that savvy players are now able to take analytical advantage of loopholes in their competitors’ gameplay. Recounting the now-legendary loss of unknown online poker player “Isildur1”, Siler suggests how Isiludur1’s victor, Brian Townsend, poured “over hand histories, creating a database of 30K hands to delineate the ranges of hands Isildur1 plays in certain situations”. Using this information, Townsend helped Brian Hastings, another pro-poker player, win $4 million from Isildur1 during a “nosebleed” session (a “super high-stakes game”) even though the “competitive edges” were “presumably small”. Internet poker has meant a shift not only in the way competitors can take advantage of others, but also in strategy. As Siler proposes;
“A book like Caro's Poker Tells (which discusses common poker “tells” which can be used to the advantage of players) may have been invaluable twenty years ago, and might still be useful if you want to play a lot of live poker, but I'd be inclined to argue that a player would be better off studying game theory, mathematics and strategy on the better online poker forums.”
PKR Poker Room
On the topic of tells, we asked Siler about PKR.com, an online poker site which incorporates the use of 3d graphics as well as social elements. Wildly differing from 2D style poker rooms, PKR.com has set itself apart by giving players the ability to choose from a variety of bodily gestures, ranging from winking and yawning to twirling their poker chips. While the incorporation of “tells” may be seen by some as a gimmick, Siler suggests that all poker sites, including PKR.com, offer players the ability to find out another’s “click tells”, with “rapid responses presumably representing strength, and slow responses representing deliberation”. The speed of the player allows others to assess the “tell”, as, quoting poker author Caro, “weak means strong and "strong means weak”.
Similar to the online casino industry, online poker has attempted to incorporate all of the Las Vegas style perks in the form of complimentary points. In the case of PokerStars.com – another popular poker website, players are encouraged to play a high volume of hands in order to receive better benefits. Such perks often cloud a player’s perception when it comes to wins/losses, as even though they’re losing money, players can find solace in the amount of comps or benefits they’ve received. Referring to PokerStars, Siler suggests that he has “seen players declare that their monthly/yearly goals involve moving up PokerStars perks levels before increasing your bankroll”. Using the example of “FGators”, a notorious player and member of the 2+2 forum (TwoPlusTwo.com) Siler observes that he was “known as a mediocre major mid-stakes multi-tabler who was a small net loser and complained quite a bit on the internet about his luck, but he also vowed to soldier on and buy a Porsche using Stars Frequent Player Points”. “FGators” ability to frame losses in view of more abstract wins has confused many other players, notably the author of a topic on FlopTurnRiver.com who titled his post on the PokerStars player as “FGators. Douche or Shark?”.
While Siler’s study offers a sociological analysis of online poker, the scientist is a gambler himself and therefore knows how players feel when their losses are racking up against them. In order to combat the “counterintuitive incentive structure” he has uncovered, Siler recommends that poker players should realise “that often you have to sometimes ‘lose the battle to win the war.’” Turning the structure on its head, Siler suggests that you can take advantage of his findings by allowing players to claim the small pots “then fleece them when you have a decent hand.” Siler also offers advice to newcomers to poker, recommending that once they have become comfortable with more mathematics based games such as Limit Texas Hold’Em, they should “expand” their knowledge of the game using books such as Small Stakes Hold'em and begin their “journey up the ladder”. Once your skills have developed, Siler recommends joining a poker room “through an affiliate that gives you rakeback and deposit bonuses”.
Siler’s study could not only influence the win/loss perception of your average poker player, his findings could also help progress the development of a No Limit Hold’Em bot. Even though Siler remains notably modest when asked whether his findings could aid the work of computer scientists, suggesting that his “research might be of some small help”, Siler believes “that advances in hardware, software and ingenuity will make a very competitive bot some point in the not-so-distant future”. In a lesser discussed element of his work, Siler also found interesting comparative data on the nature of pairs based poker hands, observing that “for small-stakes players (NL50), small pairs (22-77) were more profitable than medium pairs (88-JJ), despite the fact that medium pairs are better hands”. Accounting this data “to the fact that unskilled players have difficulty reading the ambiguous value of middle pairs”, this relatively small segment of Siler’s study shows that the scientist has a lot more to give to the online poker world. If we were to make a prediction on the future of Siler’s study, we wouldn’t be surprised to see the social scientist being offered a role at any of the online poker industry’s key businesses, as his current and consequent research will no doubt prove incredibly fruitful to poker players and bosses alike.
Casino Online wishes to thank Kyle Siler for particpating in our interview. Kyle Siler image is courtesy of Robert Barker - Cornell University Photography.