Quantcast

«

»

Sep 26 2013

A Terrible Study on Poker

Sean Carroll (the physicist, not the paleontologist) calls attention to an absolutely ridiculous study on whether poker is a game of skill. It was published in the Journal of Gambling Studies, which apparently doesn’t bother with that pesky peer review stuff. I’m not paying $39.95 for the whole study, but the abstract alone tells you how absurd the methodology is:

Due to intensive marketing and the rapid growth of online gambling, poker currently enjoys great popularity among large sections of the population. Although poker is legally a game of chance in most countries, some (particularly operators of private poker web sites) argue that it should be regarded as a game of skill or sport because the outcome of the game primarily depends on individual aptitude and skill. The available findings indicate that skill plays a meaningful role; however, serious methodological weaknesses and the absence of reliable information regarding the relative importance of chance and skill considerably limit the validity of extant research. Adopting a quasi-experimental approach, the present study examined the extent to which the influence of poker playing skill was more important than card distribution. Three average players and three experts sat down at a six-player table and played 60 computer-based hands of the poker variant “Texas Hold’em” for money. In each hand, one of the average players and one expert received (a) better-than-average cards (winner’s box), (b) average cards (neutral box) and (c) worse-than-average cards (loser’s box). The standardized manipulation of the card distribution controlled the factor of chance to determine differences in performance between the average and expert groups. Overall, 150 individuals participated in a “fixed-limit” game variant, and 150 individuals participated in a “no-limit” game variant. ANOVA results showed that experts did not outperform average players in terms of final cash balance. Rather, card distribution was the decisive factor for successful poker playing. However, expert players were better able to minimize losses when confronted with disadvantageous conditions (i.e., worse-than-average cards). No significant differences were observed between the game variants. Furthermore, supplementary analyses confirm differential game-related actions dependent on the card distribution, player status, and game variant. In conclusion, the study findings indicate that poker should be regarded as a game of chance, at least under certain basic conditions, and suggest new directions for further research.

Uh, yeah. Let me suggest one “new direction” for further research: Play more than 60 hands, for crying out loud. And don’t “manipulate the card distribution” to “control the factor of chance.” Understanding the factor of chance is one of the key elements that allows more skilled players to win over less skilled players. Bad players don’t understand probability or they ignore it to “go with their gut.” And in the short run — like 60 hands — doing so can still be profitable. Over the long run, say 6000 hands, it won’t be. Sean has some additional analysis:

So let’s confine our attention to “decision games,” where all you do is sit down and make decisions about one thing or another. This includes games without a probabilistic component, like chess or go, but here we’re interested in games in which chance definitely enters, like poker or blackjack or Monopoly. Call these “probabilistic decision games.” (Presumably there is some accepted terminology for all these things, but I’m just making these terms up.)

So, when does a probabilistic decision game qualify as a “game of skill”? I suggest it does when the following criteria are met:

1. There are different possible strategies a player could choose.
2. Some strategies do better than others.
3. The ideal “dominant strategy” is not known.

I might quibble with that last one a bit. I’d put it differently: Where the ideal strategy changes constantly, allowing the more skilled players — those who can adjust to changing conditions — to win over time. I have a friend I’ve played poker with for more than a decade and he truly thinks that there is a single right way to play poker, that it’s all strictly about probability. He’s wrong. The right way to play is going to change every time you sit down at the table, it’s going to change every time a new player joins the table, and it’s going to change multiple times over the course of a few hours even if you’re playing with all the same players. There’s a right way to play every hand depending on a dozen or more factors and the more accurately a player analyzes those factors hand after hand, the more likely he is to win over time.

Poker is really about good decisionmaking and the player who makes the right decision the most often wins in the long run. In the short run, even making the perfect decision under the circumstances doesn’t mean you’re going to win. If you get a guy to put all his money in as a 4-1 underdog, you played it right and he played it wrong even if he hits the card he needs to win that pot. And if you do that consistently, you’re going to win and he’s going to lose. That’s why it’s a game of skill.

19 comments

Skip to comment form

  1. 1
    Modusoperandi

    “Poker is really about good decisionmaking and the player who makes the right decision the most often wins in the long run.”

    Really? I just wait until I’m losing, point to the winningest player and yell, “I reckon that card came from yer sleeve! Yer a cheat!”. Then I flip the table over, hide behind it and wait for the shootout to end, picking up all the chips I can grab while I’m waiting. By the time the piano player starts up again, I’m generally up a good amount.
    This and other techniques are covered in my new book, “How to Win at Card Games, Real Old Timeylike”.

  2. 2
    scienceavenger

    60 trials? Really? I can’t be bothered to go past that, it’s Statistics 101 stuff. For anything with a small percentage advantage due to skill, and that’s almost by-definition what we’re dealing with here or there wouldn’t be a debate, 10,000 trials is just getting started.

    One fundamental problem is that the study treats “game of skill” as binary, when in reality its is a matter of degree. Tic tac toe involves some skill, yatzee more, backgammon more than that, Axis and Allies more so, chess even more than that, and perhaps Go at the top. So they are posing the wrong question: It’s not “Is Texas Hold ‘Em” a game of skill, but rather “How much of a game of skill is Texas Hold Em?”

  3. 3
    doublereed

    ANOVA results showed that experts did not outperform average players in terms of final cash balance. Rather, card distribution was the decisive factor for successful poker playing. However, expert players were better able to minimize losses when confronted with disadvantageous conditions (i.e., worse-than-average cards).

    This is blatantly contradictory.

    Obviously if they were able to minimize losses then skill matters, and it is significant.

  4. 4
    Dalillama, Schmott Guy

    Although poker is legally a game of chance in most countries, some (particularly operators of private poker web sites) argue that it should be regarded as a game of skill or sport because the outcome of the game primarily depends on individual aptitude and skill

    It’s worth noting, too, that many of the skill aspects of poker are removed in the case of video poker, which is much more probabilistic and less skill-based than in-person poker, due to a large amount of the skills of poker playing involve reading the other players as well.

  5. 5
    WithinThisMind

    Poker is as much about playing the other players as it is playing the cards. I’ve seen my husband win consistently at Texas Hold-Em without ever having better than a Q-10 off suit the entire night because he knew how to read the other players at the table and could with a surprising degree of accuracy, tell them exactly what cards they had in their hand by what was on the table and how they responded. Granted, these are his friends that he has played against for years and he doesn’t do nearly as well against strangers, but still.

    I consistently beat this man at chess every time we play, and yet I can guarantee you if we sit down to poker and it goes more than a couple hours, he’s going to come out ahead.

    Luck has it’s part, and I’ve seen many a great player screwed by the river, but hey, that’s why they call it ‘gambling’. An educated guess is still a guess, it’s just waaaaaaay more likely to be right than an uneducated guess.

  6. 6
    Reginald Selkirk

    (the physicist, not the paleontologist)

    Also not the evo-devo geneticist.

  7. 7
    colnago80

    Re Reginald Selkirk @ #6

    I think that Prof. Sean Carroll of the Un. of Wisconsin is actually an evolutionary biologist, not a paleontologist. There are only 2 Sean Carrolls in science.

  8. 8
    scienceavenger

    …could with a surprising degree of accuracy, tell them exactly what cards they had in their hand by what was on the table and how they responded

    I’ve often wondered how much of this lends itself to confirmation bias. A lot of poker talk sounds spookily like psychic talk, and we know what happens to their abilities when controls are applied. I mean, do his friends really show him their cards every hand? If they do, I wouldn’t think they were very good players.

    Luck has it’s part, and I’ve seen many a great player screwed by the river, but hey, that’s why they call it ‘gambling’.

    That’s also why a lot of us don’t think Hold Em is as much a game of skill as many of its practitioners would claim. There are no Chris Moneymakers (a relatively untalented player who managed to win the WSOP Championship) in chess.

  9. 9
    rabbitscribe

    I don’t know a damn thing about it, but when has that ever stopped anyone from posting on the Internet?

    If skill was not a significant factor in winning at poker, there would be no professional poker players because on average they’d have no more money at the end of the year than they had at the beginning;

    There are astonishingly wealthy professional poker players;

    Therefore, skill is a significant factor in winning at poker.

  10. 10
    sbuh

    Tic tac toe involves some skill

    Quibbling, but Tic Tac Toe is a mathematically solved game, and a trivially simple one to boot. It should never end in anything but a draw. Games like Chess with no random factors are solveable, we just haven’t done it yet. Too many possibilities to calculate. But in theory it should be doable in the future.

    Games with random (or approximately random) factors, things that involve cards or dice, are more probability-based, as Ed says. That plus hiding some information from the player (like the cards your opponents are holding) mean that you can never know the single correct move in every instance. You can merely calculate the probability of a correct play. And that’s not even touching on bluffs, tells, or false signals, which add yet another dimension to an already complicated game.

  11. 11
    Reginald Selkirk

    I think that Prof. Sean Carroll of the Un. of Wisconsin is actually an evolutionary biologist, not a paleontologist. There are only 2 Sean Carrolls in science.

    Well, if you don’t want to be cryptic.

  12. 12
    roggg

    Forget the 60 trials issue…the ability to minimize losses in disadvantageous situations is attributable to skill, but does not make skill a factor in successful play? WTF? I think the author has a fundamental misunderstanding about how gambling works at all.

  13. 13
    Konradius

    I actually have a dual approach to the question whether poker is a gambling or skilled game.
    First of all, I don’t think that’s an either/or question. You can have games with skill involved that still have a large gambling component in them. If you know your horses better than the people picking the odds, horse gambling might be such a game.
    For poker I would take the approach that tournament poker is a game of skill, but sit-in money poker is gambling.
    With tournament poker you know beforehand what your maximum loss is: the money you need to lay in to enter the tournament. (disregarding rebuys for ease of talking)
    But in a money game you have a true gambling situation every pot you play. And I think this would be the type of game that’s most addictive.
    Of course, real data is needed, but when you consistently fail to land in the money in (sufficiently big) tournaments that would dampen the addictiveness because the winning high is never coupled with really having won something.

  14. 14
    Michael Heath

    I’m not a poker player. I was a golfer. I compare the two because the odds any one golfer will win a particular tournament are well below 50%. There are a few outlier years where one player might break that rule, but it’s relatively rare. In fact the odds the top five golfers in the world will place in the top five for a given tournament where they all compete approaches zero, I’m not sure I’ve ever seen such happen. So some ‘good fortune’, i.e., luck, is definitely part of the game of competitive golf, even at its highest levels.

    It would be interesting to differentiate the degree to which luck plays in both games. From my perspective, amateurs have a far better chance to win a tournament in poker than they do in golf, where those odds are astronomically low. And the amateurs who do well in some golf tournaments are typically top ranked ones on the path to being a pro, and not mere laymen. So relative to golf, I think poker earns more bias about being more about luck than a sport where luck is definitely part of the outcome. Though in the long run, it’s obvious the pros in poker will enjoy far more success.

  15. 15
    Al Dente

    I can remember losing a poker hand to a guy who drew to an inside straight and got the card he needed. He bragged about that for the rest of the evening while I was taking all of his chips because he continued to play that dumb.

  16. 16
    macallan

    Probably applies to most non-trivial card games that involve some sort of strategy. In .de, the obvious example would be Skat – a skilled player ( vs. less skilled players ) won’t win every game but over time it’s pretty much guaranteed he’ll come out ahead. And there are situations where playing the players is more important than playing the cards ( ever won a Ramsch with all four jacks? Can be done. )

  17. 17
    sc_770d159609e0f8deaa72849e3731a29d

    There are at least two different sets of skills involved in poker. One is mathematical- the ability to do mental arithmetic quickly enough to assess whether the odds are in your favour or not- the other is psychological- the ability to judge what cards other players are likely to have and how they will respond to your actions and those of other players. Bluffing per se isn’t actually important; knowing players may be bluffing is- it’s when the two sets of skills clash. Never bluff and you won’t win much when you win. Bluff too often and you’ll lose a lot and more often.

  18. 18
    rickdesper

    Well, duh, of course card distribution matters. You don’t need to bring N down to 60 and feed players a skewed distribution of hands to know that. Any half-wit can win a lot of hands if he continually flops a full house.

    Presumably the result that would make them happy (or, rather, would support a hypothesis that skill was a factor) would be if all three “experts” clearly dominated the three “amateurs” getting the corresponding quality of hands. I don’t know what else they could expect to see. But with N=60, that’s really asking for a lot.

    Given how long it takes to write a paper, couldn’t they have bumped up the number of hands a bit?

    Perhaps repeated the trial? Played with the parameters a little bit?

    The most a negative result might imply is that they have failed to demonstrate that the players they selected differ from each other significantly. There’s no way that this framework is going to prove a negative result in the way they imagine it will.

  19. 19
    democommie

    I used to study poker, but the tuition was too high.

Leave a Reply

Switch to our mobile site