Steve Levitt plays poker with the FT
“I used to play poker a ton and then I quit. It’s too time consuming and toooo boring.” There’s something boyish about the way Steve Levitt drags out the word. But then his inner economist reasserts itself: “What you come to realise about poker over time is that the ratio of luck to skill in the short term is too high to make it feel productive.”
Here’s what you need to know about Levitt. He used to be a rising star in academia, with prestigious positions at Harvard and then Chicago. He picked unusual topics: cheating sumo wrestlers; the link between legal abortion and falling crime. His detective work with data was influential. In 2003, when he was just 35, Levitt won the John Bates Clark medal, often a precursor to the Nobel memorial prize. The journalist Stephen Dubner profiled him in The New York Times Magazine; a book deal followed for the pair, and the result, Freakonomics, sold four million copies. So, I’m playing poker with a data-savvy millionaire genius, a game I understand only in the sense that I’ve written about it. The good news is that Levitt doesn’t play any more. The bad news is that on his last outing, five years ago, he was within one hand of the final table at the World Series of Poker … I am doomed.
We’re at a casino in Mayfair: just me, Levitt, and the dealer, JD. At 47, Levitt has greyed since I first interviewed him nine years ago. But he still looks young and he’s better dressed than he used to be, in a silver-grey jacket and a midnight-blue shirt. JD, who deals for the poker professionals on late-night TV, looks the part in a black suit and waistcoat. Your correspondent has just come from a radio studio and is dressed accordingly. The game is Texas Hold’Em, the modern standard for poker, in which each player constructs a hand from his two concealed cards plus five communal cards on the table. The stakes: £100 each, winner takes all.
Like any good economist, I understand how to play poker in theory but am not sure how to do it in practice. (It takes me a couple of dry runs to figure out whose turn it is to bet.) We have 10,000 chips each and I have a king in my very first hand. The “flop” of three communal cards reveals a second king so, after a couple of small raises, I go in hard with a bet of 2,000 chips. Levitt chuckles, which is unnerving. After pausing, he folds. I get the impression he’s not convinced of my expertise – but I’ve won the first pot, even if it is tiny.
I’m trying to write down all the hands for posterity but that quickly becomes ludicrous. So, too, is the idea of conducting an interview while playing. Concentration is required – from me at least. I guess that Levitt wouldn’t break sweat if he had to play and chat simultaneously. I fold, and Levitt opts to show me his cards. I ask what the thinking is behind showing me that he was bluffing. Both JD and Levitt rush to explain that he had two pairs, and wasn’t bluffing at all. I realise that I have no idea what’s happening. This could be a long afternoon. Or, more likely, a short and expensive one.
“I think the statistics of poker are actually probably overrated,” Levitt says. “Most of poker is based on pretty easy rules of thumb. In a game like this there’s not many hard calculations to do.” He tells me about some research he conducted in Las Vegas, with a range of poker players including 18 winners of World Series events. “Almost all of them continued to use the rules of thumb that you use in regular poker. Even though they were not the right rules to use in the game we ran.” He concludes that experts can quickly be undone. “If you change the rules or the incentives, they tend to do very poorly.”
As our appointed break time approaches, Levitt’s getting into his stride. He’s more aggressive than I am, pushing me out of hands. But when I do stay in until our first showdown, I lose: it’s a pair of fives against Levitt’s pair of sevens. There have been no dramatic moments, yet I am slowly bleeding chips. JD is encouraging. “The play hasn’t been that bad,” he says. Levitt agrees. Still, I am losing.
Then, the very last deal before the hour, I have a decent hand: two pairs. The pot’s not a bad size and Levitt might be drawing to get a flush, so I decide to shut things down: I bet big. Levitt folds, and, as we break, I’m not far behind, with 8,900 chips to Levitt’s 11,100. I observe that since I first met Levitt, he has become a celebrity. He snorts. He’s relieved that nobody ever recognises him because he looks “so generic”.
“The nice thing is, the perks that come with the success of our book are opportunities. People come to me all the time with great opportunities.” Such as what? Money? Secrets? Power? For Levitt, the answer is simple: fun. That could mean anything from a round of golf at Augusta to working with the US Defense Advanced Research Projects Agency to prevent sex trafficking. He designs algorithms to catch credit-card fraud, and for horseracing syndicates. “The horseracing is the most fun thing.”
At one point, Levitt talks about his academic career in the past tense. “I view everything I do as a hobby now,” he says. “I no longer feel like an adult. I feel like what has happened is that I’ve been given so many opportunities that I am somehow back into a very childlike phase that I’m in the candy store and I get to pick and choose whatever I want.”
Immediately after the restart, I’m dealt a 9-7; it’s trash. But Levitt doesn’t raise the stakes so I stay in and see the flop: 4-6-8. Now either 5 or 10 will give me a straight; I call Levitt’s diffident bet. The next card is the 5. I do have a straight. It’s a monster hand in two-player poker. With 4, 5, 6 and 8 on the table, Levitt might have a straight too. But I know something he doesn’t: I have a 9, so my straight will be higher than his. Levitt comes in with a solid bet, I raise, and he calls. Then the final card comes: another 9. That’s annoying because it might allow Levitt to split the pot with me. If Levitt has the 10 and 7, he will beat me. But that’s vanishingly unlikely. I go all in. Levitt calls.
If I win, I’ll be 18,000 chips to 2,000 chips ahead. If Levitt wins, game over. And … he has the 10 and 7 of spades. I’ve lost it all. Just like that.
“He called with the nuts,” says JD. JD and Levitt are quick to commiserate. Levitt had no idea that I had him beaten all along. JD admires how I reeled Levitt in. Levitt says that I was incredibly unlucky; that last card, the 9, killed me. And the 10-7 was the only combination in the deck that could have beaten me.
“Now that’s a hand worth writing about,” says JD. I’m feeling pretty good: I’ve lost to a “bad beat” in true poker-pro fashion. But gradually the congratulation fades into criticism. Levitt points out that I should never have gone all-in. It was a small risk but a pointless one, because Levitt would never have called me except in the unlikely event that he had the 10.
“That’s an essential rule of thumb you need to know,” Levitt says. “But that’s not Poker 101. That’s Poker 403. That’s Master’s Level Poker. PhD level Poker.” Maybe. But I’ve been schooled.