The New York Times (this may cost you one of your monthly alloted articles) reports on a new article by Levitt and Miles that argues that poker is a game of skill and not a game of chance. The policy-relevant implications for the study deals with legalization – several prohibitions against poker are codified with terminology that prohibits gambling and randomized lotteries while allowing for skilled based gaming. The authors track a set of known players from the 2010 World Series of Poker (WSOP). The study indicates that "good" poker players do get returns on their tournament buyins (seen as an investment) while the unknown players are more likely to lose money during the event. The data includes 57 tournaments that run during the course of the WSOP (there are several side tournaments besides the main event).
Naturally, for those who are concerned about the endogeneity between the tracked players and their 2010 performance, the players are picked based on pre-2010 WSOP data – they were either ranked highly previous to the tournament (in poker magazines or online) or did well at the 2009 WSOP. Naturally, if poker was entirely based around randomization, your previous performance ought to have no influence on your next tournament. For those of us who are poker players, the results should not be too surprising; however, having evidence of skilled players doing well should be reassuring. Perhaps, also, it will allow us to hear less bad beat stories (stories about opposing players getting lucky against the story teller).
As an aside, footnote 10 indicates who not to be at the WSOP: "The biggest loser in the 2010 WSOP paid in $252,000 in entry fees, but earned only $24,000 in prize money."
Abstract:
In determining the legality of online poker – a multibillion dollar industry – courts have relied heavily on the issue of whether or not poker is a game of skill. Using newly available data, we analyze that question by examining the performance in the 2010 World Series of Poker of a group of poker players identified as being highly skilled prior to the start of the events. Those players identified a priori as being highly skilled achieved an average return on investment of over 30 percent, compared to a -15 percent for all other players. This large gap in returns is strong evidence in support of the idea that
poker is a game of skill.
The NBER paper can be found here.
But how does this reflect in any way on on-line poker? Presumably much of the skill is reading cues from other people’s behaviour – that behaviour, of course, includes their actual actions within the game, but it’s also other things, like body language and what-not. Arguing that real-life poker isn’t random (duh) and therefore on-line poker isn’t random is a bit like arguing that real-life fencing isn’t random and therefore dice-based D&D battles aren’t random…
It partly comes down to what we buy as a decent metric (or proxy of that metric) that evaluates skill. I buy the authors’ argument that if previous performances can predict future results, then poker is a game skill. The alternative, of course, is the game is rigged, but I think we can safely argue that this is not the case.
In the online world, the results are replicable. Sites track the performance of players and we have a lot more data. For example, a cursory look at pocketfives.com suggests that there are people who are consistent in their performance. If we really want to retest their arguments, we can use that data and look at future tournaments.
To provide a bit of anecdote, I have played poker both online and offline quite a bit and have studied “tells” in both situations. Generally, I would not suggests that tells are the primary place where the skill comes in – as players get better they are better able to disguise their tells or send false signals. So, often, the game of reading tells becomes the same as guessing what their bet really means. In the online world, you get some similar tells as well. How long a player takes to make their decision, what size their bet is, how many tables they are playing on at the same time, and whether or not they are chatting in the box can all give you some indication of their mental state and what their bets may or may not mean.
Personally, I have usually done better in person than online, but that was not because of the lack of tells; it was mostly due to the player types. In person players are more likely to be casual, visiting a casino for the weekend, old school style of playing poker, or an average poker player. These styles are generally easier to play against. Online, especially starting about 2003, players became more aggressive. This was generally attributed to younger players becoming involved in the game, but also the high variance strategy of aggression on multiple hands is able to force other players to fold or give way to bad reads to the opposition. The younger, aggressive players do play in the casinos, but there seems to be a smaller percentage of them in person.
I think every game is a skill whether it involves money or not..because skill can be achieve or enhance through experience so in every game you must have a experience to have a skill in that game just like playing poker..
Very informative article.Much thanks again. Awesome.