This is the 16th installment of Viewfinder, my weekly column for Yahoo! India, and was published on August 12.
I spend as much time playing poker these days as I once would spend reading and writing, and my friends sometimes ask me in jest what literature and poker have in common. My reply is that both provide an understanding of human nature. I am not being facetious.
Ever since I started playing poker seriously, I’ve held the view that poker reveals the way our brain is wired. For example, if we carry a list of cognitive biases with us to a poker session, and tick off the ones we witness in action, we’d probably run through the entire list by the end of the evening. If we’re aware of this, we can exploit these missteps in others—and avoid them in our own play.
In writing this article, I run the risk of revealing to my regular opponents a few of the tricks of my trade. But for the greater good of humanity, I shall lay those considerations aside. Here, then, are a few of the cognitive biases that come into play on a poker table.
1. The Sunk-Cost Fallacy. Suppose you are in office one day, and there is much buzz about a new Japanese restaurant that has opened round the corner. “Let’s go there for lunch,” you suggest. All your regular cronies concur, except one girl who says, “I so want to come, but I’ve got lunch from home, and it will be wasted.” That is her only reason for not coming. She doesn’t want the packed lunch, and would vastly prefer some unagi, but the Sunk Cost Fallacy comes in the way.
The logical way of thinking about this is that the packed lunch is a sunk cost—and that if she would otherwise prefer to come to the new Japanese restaurant, then she should ignore that sunk cost and come anyway. This is the same mistake many stock market investors make. They will buy a stock for, say, Rs 70. It will slip to Rs 60. Its downward momentum will make it logical to sell the stock, but they will reason that they have already lost Rs 10 on it, and will keep the stock in the hope of recovering that money somehow.
Poker players make the same error by throwing in good money after bad. Let us say that you have pocket aces. You raise pre-flop, a loose player calls, and the flop comes AQJ with two hearts. (You have none.) You have a set, but slow-playing is dangerous because of the flush and straight possibilities out there, so you make a pot-sized bet. Your opponent calls. The turn is a ten of hearts. You make a bet two-thirds the size of the pot, and your opponent raises three times that. For any good player, unless you have a read that the opponent is weak, this is an auto-fold. There are four cards to a straight out there, three to a flush, and if your opponent has one of those, you have exactly ten outs to a full house or quads, and the odds don’t justify continuing. But you say, “I have already spent so much money on this pot. All that will be wasted. I can’t leave now.”
Good poker players know that the money already in the pot no longer belongs to you, and that at every street you must make new evaluations about how to proceed. But we are human, we have put money in the pot, and it’s so hard to let it go. Isn’t it?
Also see: Escalation of Commitment.
2. The Endowment Effect. The above poker example also illustrates the Endowment Effect, which Wikipedia describes as “a hypothesis that people value a good or service more once their property right to it has been established.” It’s been much written about recently in a slew of books about behavioural economics, and is a bias we often see in poker when a player ‘falls in love with his hand.’ In the above example, if you are a spectator watching the hand, it is obvious that the set of aces should be folded. In the middle of the action, though, you ascribe more value to the hand than you would if some other player held it because it’s your hand, and it’s so hard to let it go. Almost all regular players have faced a situation where they play AK, flop top pair-top kicker, but their bet on the flop encounters a big raise (or even an all-in) from a solid player who doesn’t make crazy moves. Seen from the outside, it’s time to consider folding, because he could have a set or two pair, but if you’re the guy holding AK, it’s so much harder to make that dispassionate decision.
When I started playing poker, I’d refer to this as the Starting Hand Bias. Weak players who hold JJ will often be reluctant to fold to a bet following a flop that has two overcards, and players who have AK or AQs will find it hard to give it away when they don’t connect on the flop. It takes discipline to overcome this bias and throw the hand away.
3. The Normalcy Bias. Wikipedia defines this as “an extreme mental state” that “causes people to underestimate both the possibility of a disaster occurring and its possible effects.” This is related to the Availability Heuristic, “a phenomenon in which people predict the frequency of an event, or a proportion within a population, based on how easily an example can be brought to mind.”
Two examples come to mind from my own play, against the same opponent. In one case, there were four cards to a flush on the board, with no repeat cards, and I had the ace of that suit—in other words, the nut flush. But the four cards were connected with a gap in between, and there was the small chance that my opponent had the one card that made her a straight flush that beat my hand. I raised, she insta-reraised, and my read was that she was very strong. But I thought, “Nah, straight flushes are so rare, she can’t possibly have one.” I did refrain from re-reraising all-in, though, and merely called, to be shown the only hand that could beat mine.
In another hand, I had a full house and was reraised on the river. The only hand that could be beat me was quads, and my opponent, who is not difficult to read, showed immense strength. Quads are so rare, though, that I ignored my read and called. You guessed it: Black Swan event.
We see the same phenomenon when a player flops a low flush, and is quite happy to reraise all-in, assuming that his hand is surely the best hand, because hey, he can’t remember the last time two players flopped a flush. That’s exactly the kind of hand that busts players out of tournaments.
4. The Recency Effect. This can be defined as “the tendency to weigh recent events more than earlier events.” Wikipedia gives an example: “If a driver sees an equal total number of red cars as blue cars during a long journey, but there happens to be a glut of red cars at the end of the journey, they are likely to conclude there were more red cars than blue cars throughout the drive.”
In poker, this can lead us astray against loose opponents. Let us say that in the last half an hour of a session, you have seen a player raising with KQo, QTo, 79s, A6s and even 58o, all marginal (some outright dubious) hands, especially from early or middle position. So you’re in a hand where he’s raised from early position, and you have AJs. You reraise, he calls. The flop is A23 rainbow. He checks, you bet the pot, he reraises by three times, a move that recent evidence indicates he is capable of making with nothing. What do you do?
I’ve gone all-in a similar situation, only to be shown AK. I had fallen prey to the Recency Effect. I’d made a move based on his recent play, quite ignoring that even loose players get good cards, and that my hand, because of the jack kicker, was not quite a monster.
This is a bias that good players can exploit successfully by changing gear in the middle of a session. Play loose for a while, then suddenly go tight, and you will get paid off on premium hands. Play tight for a bit, and then make a bluff, and your opponents will give you more credit than is due and fold.
Also see: The Primacy Effect, “the tendency to weigh initial events more than subsequent events”. You often see sharks exploit this by starting a session with some loose play, for advertising effect, so they get paid off on their premium hands later by players overvaluing marginal holdings. In other words, these sharks behave like fish at the start of a session, and later go chomp chomp chomp.
5. The Confirmation Bias. This is the tendency to ignore all information that contradicts our preconceptions, and to treat all other information as evidence. People who believe in astrology, for example, will remember all the instances when an astrologer’s predictions came true, and ignore all the times they did not. Ditto homeopathy, and suchlike.
I see this all the time with poker players. I know players, for example, who love to play hands like 58o and 63, and will call big preflop raises with them. They have stories about how they once flopped a straight with 63, beating two opponents who had AA and QQ, and so on. Another player I know has a goofy theory that if two or three players have shown strength with preflop raises and reraises, and he has two low cards, he should call because the other all surely have high cards, so there is a greater probability of low cards hitting the board. (Go figure.)
Players with beliefs like this remember the handful of times such play works for them, and ignore all the other times when it doesn’t. If you play a hand like 85o, you will flop two pair or better approximately one in 34 hands. The rest of the time, you are basically losing money. Weak players remember the one time they hit—not the 33 times they don’t.
Also see: The Semmelweis Reflex.
* * * *
This is a subject on which I could go on and on: there’s no end to the cognitive biases one sees at a poker table, from Loss Aversion to the Choice-Supportive Bias to the Ostrich Effect to the Belief Bias and obvious ones like the Optimism Bias, the Over-Confidence Effect and the Neglect-of-Probability Bias (duh). Check out this list of cognitive biases at Wikipedia: if you are a poker player, you will surely recognise many of them.
For a while now, I’ve been mulling over the idea of writing a book about how the game of poker reveals how the human brain is wired—so this may not be the last you hear from me on this subject.
* * * *
Previous articles on poker:
Throw a Lucky Man into the Sea
* * * *
Previously on Viewfinder
Throw a Lucky Man into the Sea
Indian Liberals and Colour Pictures
Internet Hindus and Madrasa Muslims