Author of Poker Study Shares His Thoughts on FindingsNew Report Analyzes 103 Million Poker Hands |
|
In a conference call today, Paco Hope — the technical manager of Cigital and one of the authors of the study — took questions from Card Player and further explained the report’s findings.
The following is excerpts from the conference call, edited for clarity and brevity:
Paco Hope: There are several laws in the U.S. and elsewhere based on this idea of a dominance test. They view gambling based on how random the game is; if randomness predominantly determines the outcome of the game, then it’s considered gambling. Otherwise, it’s not.
Plus, there’s the public perception that poker is essentially a game of random cards. The study really goes to the heart of that by looking at real numbers, at what do we actually see on online poker — which is pretty representative of poker generally, because there are the same rules, it’s real money, it’s real players.
So, what we did is we got 103 million hands of poker logs from hands played in cash games — not tournaments and not play money — from PokerStars.
Did the hand actually have a showdown? If it didn’t, the cards didn’t pick the winner in that game. It was the players’ decisions, all the players decided to fold … essentially they’re exercising skill over the game.
If there actually was a showdown, did the person with the best five-card hand according to the standard poker rules actually win that showdown? And that only turned out to be half the time. Half the time, the person who would’ve made the best five-card hand in the end actually folded earlier. So, the point that we take away from this is that the vast majority of the time — 75 percent of the time — decisions completely determine the outcome. Everybody decides to fold, one guy takes the pot. In the remainder, half of that time, it is the decisions of the person who held the cards.
Poker Players Alliance used this study amongst other things when they filed an amicus brief in a recent court case in South Carolina.
Card Player: How did PokerStars get involved with this?
PH: PokerStars initiated this survey. They asked us to conduct it and they provided the data, but all they did was say, ‘We want you to do this analysis,’ and then they stepped back and said to tell them what we wanted. I was given free rein to pretty much ask them for anything that I wanted …We agreed on the 100 million hands as a good, significant number.
CP: So PokerStars funded the study?
PH: They did. But what we wanted to do was establish some level of independence. Obviously, we didn’t want people to say, “Oh, they funded the study. They could give us a bunch of data that tells us what they want us to say.” What we tried to do was verify what they gave us. They gave us a big log file, so we went out to twoplustwo, which is a very popular online poker forum, and said, “Hey gang, send us your hand histories so we can corroborate what PokerStars sent us.” Out of 103 million, I think there was around 700,000 hands that we corroborated. I realize that’s only a small fraction, but for those, they were exactly what PokerStars said they were.
You will probably see more of these studies along these lines in the future. We’re talking about doing something with the Sunday Million tournament that PokerStars runs every Sunday, getting a few years of that data and running the same analysis for people who argue that tournament play is something else. We’re going to do the same analysis to see if it comes up the same way.
CP: Besides your two published findings, was there anything else you were looking for or anything else that stood out?
PH: We found a few interesting things that I can talk about generally. When you look at that number — 103 million hands — there’s a lot of nuance that’s lost based on the number of players at the table or the limits at the table. At a 10-person table with 10 cent stakes, you see a lot more showdowns at a low table like that than you do at a $100 no-limit game with three people at the table.
That was something that we didn’t report on specifically, but it is interesting to note the way the showdown percentages dropped like a rock the lower the number of people you have and the higher the stakes. So, high stakes, at a three- or four-handed table, you’re going to see showdown rates in the 20 percent range. Eighty percent of the hands, roughly, are decided by the decisions of the players. And then the showdown rates could be in the 50-60 percent rate.
CP: Do you think you might publish additional findings from that data that may not be specifically relevant to an overall legal argument but would be of interest to poker players?
PH: Well, we have to be careful with it, because some of it is regarded as proprietary. I know that a professional poker player has a voracious appetite for information, anything that might improve their game or give them insight into the game. So, I know that there is a demand for it. But we have to be careful what things are considered proprietary.
So, most of the information that I release publicly, I have to vet with PokerStars first. I mean, I’ve got everything. I can tell you every time a guy was bluffing and if he was bluffing, what he had. So, I have to be careful about that. I don’t know how much we can release. We might, for instance, be able to release graphs, but not numbers, so that you could see the shape of a curve but you didn’t know actual data points.
CP: This report has already had an impact, in terms of the South Carolina case. But what kind of a future impact do you see this report having on upcoming legal cases involving poker.
PH: We’re trying to create a study that people can take and say, this is an example of some actual, hardcore data that poker is a game of skill as opposed to a lot of the anecdotal evidence. A lot of times they’ll bring in expert players to the stand and have them testify how much skill is involved. We’re trying to get away from some of things that appear to be anecdotal and move it to science. We’re trying to move it out of the anecdotal and the persuasive element and move it into the nitty-gritty science element.
Copyright © 2001-2024
CardPlayer.com. All rights reserved.
