Is poker a game of skill or chance?

By Seriously Science | September 23, 2013 12:00 pm

Photo: flickr/Ulf Liljankoski

Can you really get better at poker? How much skill does it take to win? In this ‘quasi-experimental’ study, the researchers set out to answer these questions. To do this, they had both expert and novice poker players sit down to fixed games, receiving prearranged card hands. This allowed the authors to test whether skill or chance could predict the outcome of the game. The results revealed that although the cards dealt pretty much predicted the winner, skill was important for reducing losses when players are dealt a bad hand. So, if you are a beginner, you’d better have beginner’s luck or you might be broke by the end of the game.

Is poker a game of skill or chance? A quasi-experimental study.

“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.”

Related content:
Discoblog: NCBI ROFL: Epileptic seizures evoked by the Rubik’s cube.
Discoblog: NCBI ROFL: My smile beats your p-p-p-poker face.
Discoblog: NCBI ROFL: Surprising study finds that little old ladies enjoy playing bingo.

CATEGORIZED UNDER: holy correlation batman!, rated G
  • Tim Converse

    I’m not sure where to start in criticizing this study. :)

    But let’s start here: let’s assume that success at poker is due to a mixture of luck and skill.

    Next, let’s assume that “luck” evens out over time – that is, the more hands that two people play against each other, the less they will have differential luck in terms of the cards they receive.

    This implies that the right way to study the question of whether poker skill exists is to pit players of differential (supposed) skill against each other over many many random hands, where we assume that the advantages of luck will have evened out. If the “skilled” players then outperform the novice players to a statistically significant level, we should assume that that skill exists.

    This paper takes the opposite tack: it gives certain players substantially better hands over a very small number of hands (60 hands), and then asks whether the variance can be mostly explained by the cards. Of course it can, for two reasons:

    1) Better cards lead to better outcomes. The only reason that this doesn’t lead to skill not mattering is that *no* *one* gets better cards, consistently, over the long term, unless they are cheating. The differential impact of “lucky” cards approaches zero as the number of hands approaches infinity.

    2) The number of hands in the experiment is small enough that you would expect luck-of-the-draw to dominate.

    I would like to see a “replication” of this study, in one of two forms:

    a) rather than 60 hands, deal 6000 hands
    b) leave it at 60 hands, but make the luck difference between the different groups arbitrarily slight.

    Better yet, just test experts against average players on hands that are in fact randomly generated.

    • ddsouza

      In other words, the effect of skill is so small in poker that it would take literally 6,000 hands for an experimenter to discover a significant effect of skill (and a much reduced size of luck). Sounds like poker is more a game of chance than skill…! By contrast, in a real game of skill (e.g., football), if you put me in goal against Manchester United, they would score with their first shot on target and wouldn’t need the full 90 minutes to humiliate me. (I predict that a lot of adrenaline-fuelled people are gonna attack me on this lol!)

      • Tim Converse

        No, I didn’t mean that you would need 6,000 hands to see an effect of skill in a well-designed study – one that tested players of different skills with no pre-arranged difference in “luck” (i.e. with randomly dealt cards).

        Of course, how many hands it would take depends on the size of the skill difference. I believe that Phil Ivey could predictably beat me in a 60-hand test with fair cards. Demonstrating Phil’s superiority over a run-of-the-mill successful poker pro might take a lot more hands.

        As to the original study: To beat your Manchester United analogy into the ground, imagine that it’s a game of penalty kicks, where you alternate being the kicker (with a normal ball) and the defending goalie (where Man United has to kick a 500-lb ball). At the end, notice that Man United has scored no goals (and quite likely you haven’t either). So where’s the skill in football?

        • ddsouza

          Your analogy is terrible. It’s not even slightly helpful. Your analogy suggests that the researchers did this:

          1. Conducted a study with experts and non-experts.
          2. Made it literally impossible for the experts to win.
          3. Reported (unsurprisingly) that the non-experts either drew (0-0) or won.

          This is not a good analogy! It’s not even close to the actual design of the study! Ah well… I guess you already know this.

  • Kyle Heinsch

    This is a bad study. It takes thousands of hands of poker to establish a true win rate. Even under controlled circumstances, 60 hands is nothing.

    Furthermore this study is unnecessary. Internet players have databases of hands numbering in the millions, in some cases with several thousands of hands against a common opponent. Even a cursory analysis of these databases will reveal skill as a factor.

    To keep it short, in the short run luck rules. In the long run a slight skill advantage relative to your opponents will reveal itself as a positive win rate. To repeat, it takes many thousands of hands to filter out variance. What you saw in your study was simply variance due to an insufficent sample size.

    • ddsouza

      Expert poker players couldn’t outperform average players over 60 hands. I find that interesting. It would also be interesting to see how many hands (on average) experts need before the luck of the cards averages out and expertise plays a significant role. Also, I would like to see whether one or two lucky hands early on can have non-linear effects on the outcome of a game.

  • dorkyman


    If you know the various probabilities of different hands better than your tablemates, you will win long-term. Poker and Blackjack are the only games where expertise can produce profits. It will take many hundreds of hands to show it, though.

    • ddsouza

      I’m sure non-experts can work out the (approximate) probabilities of different hands. Even if they can’t, I bet many can quickly develop an “intuitive feel” for what cards are winning cards, what cards are losing cards, and what cards are unlikely to appear (e.g., that damn 4 when there’s a 2, 3, 5, 6…).

      I bet the main difference is that experts know when to fold and are not always so tempted to stay in.

      • Ebola Jesus

        Odds come into play when players try to decide whether they can profitably call a bet on a draw. It’s not the entire game. You can tell it’s not the entire game because at any point the other players can, if they so choose, slam you with oversized bets on every street such that you never have “proper” odds to call, rendering your knowledge of pot odds completely useless.

        At this point you can either fold every hand (and lose) or start firing back whenever you hold a premium hand. But you need to decide what a premium hand is based on the range of hands you think he holds when he bets into you, then decide when the flop comes how likely you are to be holding the best hand given the range of hands you think he would have called a raise with, and what is the likelyhood of him holding a worse hand that can call a bet of X size vs a better hand, and so on, and so on.

        By all means, though: go join a random 6-max game, bet whenever you think you “have a good hand” and see what happens. (Hint: you’ll win a couple big pots, lose a couple huge pots, and lose a ton of small pots.)

  • Aaron W

    “In conclusion, the study findings indicate that poker should be regarded as a game of chance, at least under certain basic conditions”

    What conditions are those? Lab conditions that don’t reflect the actual distribution of cards in a real-world game of poker?

    In a real game of poker, players get bad cards as often as they get good cards. This study shows good players lose less money when they’re playing with bad cards. So it’s a game of skill, unless a lab technician (or crooked dealer) is controlling the distribution of cards.

    I don’t see how the authors can conclude poker is a game of chance, when their findings show otherwise.

  • Dante Finder

    In no way is this a true scientific study as Kyle said. I wonder why they posted something like this as I came here from an article directory.
    I always like to say, “There’s a chance Poker is a game of skill, but there is no proof.”

  • Poker Games

    Poker is a game of skill because it is the only existing game of chance in which it is upto you to maximize your profits and minimize your losses.


Seriously, Science?

Seriously, Science?, formerly known as NCBI ROFL, is the brainchild of two prone-to-distraction biologists. We highlight the funniest, oddest, and just plain craziest research from the PubMed research database and beyond. Because nobody said serious science couldn't be silly!
Follow us on Twitter: @srslyscience.
Send us paper suggestions: srslyscience[at]

See More


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar