Episode 749: Professor Blackjack : Planet Money Ed Thorp started his career teaching math at MIT. Then he slid sideways into blackjack, changed the game forever, and set his sights on Wall Street investing. He changed that forever too.

Episode 749: Professor Blackjack

  • Download
  • <iframe src="https://www.npr.org/player/embed/510810752/510815704" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


Ed Thorp was 28 years old. He was an academic, a scholar. He'd just started teaching math at MIT. And on a grey February afternoon, a blue Cadillac pulled up outside his apartment.

ED THORP: A pretty blond got out of the passenger side. Another pretty blonde got the driver's side. And then this white-haired, kind of short, elderly man in long cashmere coat got out. He had been snuggled up between the two blondes.

GOLDSTEIN: They knocked on the door. Thorp them in. The man introduced the young women as his nieces.

THORP: I was naive, and I believed it, but my wife angrily disagreed (laughter). These are not nieces.

GOLDSTEIN: And what was this guy's name?

THORP: His name was Emmanuel Kimmel.

GOLDSTEIN: It's 1961. Kimmel tells Thorp he's a wealthy businessman in the parking lot business in New York. He does not tell Thorp that he used to be in the illegal gambling business in New Jersey. Kimmel's there because Thorp had just published a math paper claiming that he could do what nobody had ever done before. He could consistently beat the casinos in Vegas.

He'd come up with a system for the card game blackjack that he said would give the player an edge. It was just a theory. It was just a math paper, but Kimmel thought it could work in the real world. And he said to Thorp, let's practice a little bit. Let's test it out, and then let's get on a plane to Nevada and hit the casinos. You can use my money.

THORP: Well, he wanted to start with $100,000.

GOLDSTEIN: Oh, my god.

THORP: I thought to myself, you know - $100,000, by the way, is almost 10 times that amount in today's dollars. So...

GOLDSTEIN: So it's like somebody go into something math kid and being like...

THORP: Let's go to Vegas with a million dollars and play blackjack with your system.

GOLDSTEIN: Two months later, Ed Thorp was sitting at a blackjack table in Reno with thousands of dollars in chips and a new way of thinking about risk.

Hello, and welcome to Planet Money. I'm Jacob Goldstein. Today on the show - how Ed Thorp set out to be a scholar, slid sideways into gambling and went on to earn hundreds of millions of dollars as an investor. He changed the way people play blackjack, and he changed the way Wall Street thinks about making money. He was the first quant.


GOLDSTEIN: A few years before that day the Cadillac pulled up outside his house, Ed Thorp was living in L.A. with his new wife. He'd just finished his Ph.D. at UCLA, and a professor had told him about these soldiers who had spent their time in the army playing blackjack. They were mathematicians, and they wound up publishing a paper on how to play the game. So Thorp drew up a little card - like, this little cheat sheet with instructions based on their system - and he went to Vegas to try it out.

THORP: About halfway through, people were ridiculing me. They - I was playing slowly. And the dealer was saying, you know, I'd - you know, what farm did I come from?

GOLDSTEIN: To Vegas, Thorp is just one more sucker with the new way to lose money. You know, people have been coming up with systems forever. But no matter what system you use, including the army guys' system, you are still going to lose in the long run. The odds are still going to be in the house's favor.

In blackjack, it's just you against the dealer. You add up the value of your cards, and the winner is the person who gets closer to 21, which, by the way, is why blackjack is also called 21. If you go over 21, you lose automatically - even if the dealer also goes over 21. You start out with two cards. You keep asking for more, one at a time. Maybe you get three cards, maybe four. More than that is really pretty rare. So Thorp's plugging away at the table using this little strategy card - this cheat sheet that tells him how to play. And he's getting mucked, hand after hand.

THORP: And then, amazingly, the card created an unusual situation.

GOLDSTEIN: In this one hand, Thorp's little cheat sheet told him to just keep drawing and drawing and drawing. It just seemed totally ridiculous, but Thorp didn't think about it. He just did what the cheat sheet said, kept asking for more cards, wound up with seven cards - unheard of. And they added up to 21 on the nose. He won.

THORP: They were initially ridiculing me and thinking - another fool of the system, as I looked at my little card. And then when I got the seven-card 21, they seemed quite in awe of what was going on. So I said to myself, what do these people know?

GOLDSTEIN: He realized the experts in Vegas - the people who make the rules at the casinos - didn't really understand blackjack. He thought, if I could just improve the army guys' system a little bit, I could tip the math in my favor. I could beat blackjack.

THORP: And so I set out to devise a system to do just that - beat it.

GOLDSTEIN: The game had been around for hundreds of years. People had been trying to beat it, including, like, smart people - right? - who were, you know - who knew probability.

THORP: That's a good point. Probability itself got started because mathematicians were interested in trying to figure out the odds in gambling games and, in particular, how to beat gambling games. And the result was you can't beat them, so...

GOLDSTEIN: Meaning that they discovered that the house always wins in the long run.

THORP: Exactly right. Yeah. So no matter...

GOLDSTEIN: Yeah - which you don't need probability to know already.


GOLDSTEIN: So Ed Thorp is going up here against not just the Vegas casinos, but - you know, against, like, hundreds of years of the history of probability theory. He goes back to UCLA, goes to the library to take another look at that article from the army guys.

THORP: I'm standing in the stacks. I pull the article down. And I decide to start reading it while I'm standing up.

GOLDSTEIN: Staring at the article, he has this insight. Blackjack is fundamentally different than other gambling games.

THORP: Most of the games - whatever happens on one trial or one player of the game - doesn't have any influence on what's going to happen next. So if you roll a pair of dice and get a five on one roll, it has no influence on what totals are going to come up on the next roll.

GOLDSTEIN: If the roulette wheel comes up red on one spin or on two spins or on three spins, the odds of whether it's going to come up red on the next spin do not change, but blackjack was different. At the time, the dealer dealt from the same deck for several hands in a row, which meant that, say, if you got an ace in the first hand, you were in fact less likely to get an ace in the next hand. Other systems at the time, including the army guys' strategy, only looked at what was happening in one hand. Thorp thought, it's not just what's in your hand that counts. It's what happened in the hand before that and the one before that.

THORP: As I'm going down the page, I realized in a minute or two that if cards were used up during the play of the game, the odds would shift back and forth, sometimes for the casino, sometimes for me.

GOLDSTEIN: Thorp realized he needed to stop just looking at the cards in his hand and start thinking about the whole deck. And right away, he thought of this math technique he learned in grad school - it's called measure theory - that could help him work out the details.

THORP: The idea was pretty straightforward to me. The hard part was actually doing all the calculating.

GOLDSTEIN: It took him more than a year to do all the calculating. He got that job teaching at MIT. He'd go into the math building in the middle of the night to work on the calculations, started using the department's punch card IBM.

THORP: As the weeks and months went by, I finally got to the point where I could ask the machine this question - what would happen if I took all four aces out of - out of the deck? So the machine came back and said, well, an ace-poor deck hurts the player, and I can calculate from these results how much.

GOLDSTEIN: In other words, he now knows that if he's sitting at the table playing blackjack, and he sees a bunch of aces go by - they're out of the deck now - his odds in the next hand are a lot worse. He should bet as little as possible because he is probably going to lose.

THORP: I did it with all the other cards, and I found out that if the fives go out of the deck, the player gets a huge edge - bigger than any other card.

GOLDSTEIN: He figures out that if a lot of small cards go by and a lot of big cards are still in the deck, the odds are in his favor. He is more likely to win than to lose. And he calculates that if he bet small when the odds are against him and big when the odds are in his favor, he will win more than he will lose. Ed Thorp, you are the first person to develop a scientific system to beat the casinos at blackjack. What are you going to do now?

THORP: I said, well, I'm going to publish this in a scientific journal of one sort or another.

GOLDSTEIN: But, like, why? Like, why were you more interested in publishing than in making money?

THORP: What I thought was that I was going to lead the life of an academic. I thought that teaching in the university would be a great life, and I would just enjoy doing what I did. And that's pretty much all I cared about.

GOLDSTEIN: And did this seem like a step in that direction - like, you know, making you out to be sort of an interesting math researcher-scholar?

THORP: Well, I didn't think of it that way. I just thought - here's something that's going to amaze you guys.

GOLDSTEIN: (Laughter).

THORP: You haven't...

GOLDSTEIN: It's like doing a magic trick or something.

THORP: Right. Exactly.

GOLDSTEIN: The Proceedings of the National Academy of Sciences published Thorp's article. It was titled "A Favorable Strategy For 21." It got picked up by newspapers all around the country, and once again, people start making fun of Thorp, saying he's just one more crank with a system. A casino spokesman said, quote, "when a lamb goes to the slaughter, the lamb might kill the butcher, but we always bet on the butcher." Thorp wanted to prove that his system was for real - that it would really work.

So that day when Manny Kimmel drove up in the blue Cadillac and offered to stake Thorp, Thorp said yes. He talked Kimmel down from that original $100,000 offer - just seemed like too much money, made him nervous. They wound up going with $10,000, which was still more than Thorp got paid in a year. And two months after that first day, Thorp was ready to fly off to Nevada with Kimmel. But before he left, Thorp's wife asked him to call every day just to let her know how things were going.

THORP: And I said, you know, phone calls are pretty expensive. Why don't we - there's a way we can do it without having to spend any money. Here's what I'll do.

GOLDSTEIN: Every night, Thorp says, I'll call you collect. I'll ask to speak to Edward Thorp, and I'll add a middle initial. That will be like our secret code. So my middle initial is A, that means we're up zero to a thousand dollars. If it's a B, we're up 1,000 to $2,000.

THORP: And so on, all through the alphabet, then Thorp's wife refuses to accept the collect call. It's a way of sending a message for free.

THORP: And when we hit our peak - I think we're up 13,000 at one point - I asked for Edward M. Thorp, and she sounded very happy.

GOLDSTEIN: So just to be clear, you were up by more than $10,000. You had just made more than $10,000 in a few days of playing blackjack.

THORP: Yeah.

GOLDSTEIN: And you didn't want to pay for the long distance call.

THORP: Well, you know, I grew up in the Depression, and that mentality sticks.

GOLDSTEIN: Kimmel is backing Thorp, and he gets to keep most of the winnings, of course, but Thorp's share was more than a thousand dollars - about two months' salary. And this started to change the way he felt about money.

THORP: For the first time, we didn't have to live hand-to-mouth for a little while. And that was - that was a pretty nice feeling. What I thought to myself was, well, Las Vegas could be kind of a piggy bank for me. If I need something, I can just go out there and get the money and buy it.

GOLDSTEIN: A lot of people feel that way about casinos. The difference is everybody else is wrong, and you are right.

THORP: Yeah, right.

GOLDSTEIN: Thorp does go back to the casinos with his own money, does win more money. And he writes a book called "Beat The Dealer." It's a best-seller. He becomes, like, a minor celebrity, goes on a game show.


UNIDENTIFIED MAN: Will the real Dr. Edward Thorp please stand up?


GOLDSTEIN: Casinos start filling up with, you know, people who want to be like Ed Thorp, people who are trying to count cards. They start changing the rules, using more decks, shuffling more often to make it harder to count cards. And Thorp - you know, he'd never set out to get rich in casinos or whatever. He's pretty much done with casinos at this point, but he thinks to himself, you know, I beat the casinos. Maybe I can go bigger. There's this obvious, much bigger casino out there - the stock market. Maybe I can beat that. He tries picking stocks.

THORP: So I made a few investments, and they worked out terribly.

GOLDSTEIN: (Laughter) Meaning, when you say they worked out terribly, you lost some money a lot of money, a lot of money - what does that mean?

THORP: Well, I bought one stock at 40, went down to 20.

GOLDSTEIN: Yeah. Suddenly, Thorp is once again just one more sucker doing what everybody else is trying to do - in this case, figure out which stock is going to go up. He needs to figure out how to do what he did in blackjack, how to zoom out - you know, not just do a better job of what everybody else is doing, but look at the market in a whole new way. He starts reading about the market in stock options, in particular something called a call option. A call option gives you the right to buy a particular stock at a set price before some future date. In the late '60s when Thorp started looking at this, call options were seen as really risky.

THORP: In those days, it turned out that people thought of calls as lottery tickets.

GOLDSTEIN: There's a good chance a call winds up being worthless and some chance that you can make a ton of money on it. Thorp realized that the prices people paid for calls were all over the map.

THORP: They didn't know how to value them.

GOLDSTEIN: This is like that moment in the casino where he realized that the people running things didn't really understand what was going on. And again, as with blackjack, he sort of zooms out - you know, thinks bigger, uses math, does a lot of calculations - and tries to find an edge.

THORP: And I got a really neat formula.

GOLDSTEIN: (Laughter).

THORP: So I said, I like this formula. It looks right to me. I'm going to try it out.

GOLDSTEIN: The formula worked. He figured out a way he could trade both call options and the underlying stock so that he'd be protected if the price of the stock went down, and he started making money consistently. A few years later, by the way, a couple of economists came up with this same formula. And they published it, and their work went on to win the Nobel Prize.

Anyway, Thorp starts managing money for some of his friends and colleagues, goes on to start an investment firm - the first one that is really driven by math and technology. And the firm has this incredible run of using quantitative techniques to find just one edge after another. For something like 20 years, it is one of the most successful investment firms in the country.

THORP: In, I think, 230 months, we had three down months. All the others were up, and three down months were all less than 1 percent.

GOLDSTEIN: So like that's...

THORP: We had...

GOLDSTEIN: I feel like that's not supposed to happen. Nobody makes a profit every month for 20 years.

THORP: No. Right.

GOLDSTEIN: Is it because you're just smarter than everybody else?

THORP: I'd hesitate to say that. There are lots of smart people in this world.

GOLDSTEIN: Yeah. So why? Why did you do this wildly improbable thing? Why could you do it?

THORP: Well, I was lucky in that I came at investments through blackjack tables. And the blackjack tables are an amazingly good training ground for learning how to invest, how to think about investments, how to manage them. And the reason is that they teach you, on the one hand, to use probability and statistics to evaluate things. And on the other, they teach you discipline. When you find something, you stick to it.

GOLDSTEIN: The firm was called Princeton-Newport Partners. The Newport part was for Newport Beach, Calif. That was where Thorp ran the firm's strategy office. The Princeton part was for the Princeton, N.J., office. That was where all the trading happened. In the 1980s, some of the people in the Princeton office on the East Coast were convicted in federal court for this scheme to avoid taxes. A lot of the charges were overturned on appeal. Thorp himself was never charged, says he didn't know what was going on, but that was the end of the firm. The firm shut down.

After that, Thorp found a few more ways to beat the market. He bought a stake in an oil tanker when oil tankers were selling for scrap value, more or less. During the dotcom bubble, he figured out how to profit in the middle of that craziness. And in all of these deals and all of these markets, he says, he's thinking about things the same way he thought about blackjack.

THORP: Yeah. I'm looking at it in a new way.


THORP: And I'm using new information from the new way that I approach the game. It's the same with all these other approaches. If you - if you bring in more information, what seemed unbeatable may become beatable.

GOLDSTEIN: Thorp is 84 years old now. He's basically out of the investing business, but the investing business has transformed itself to become more like Ed Thorp. You know, before he came along, you didn't get a math Ph.D. or a physics Ph.D. and go off to Wall Street. That just was not a thing people did. Now, it is a totally normal thing that people do. Wall Street is now full of people trying to do what Ed Thorp did - trying to use math and computers to find new ways of looking at markets.


GOLDSTEIN: Do you have a surefire money-making Email it to us - planetmoney@npr.org - or tweet it to us or talk to us on Facebook. Ed Thorp wrote a book about his life. It's called "A Man For All Markets." It comes out later this month. Our show today was produced by Nick Fountain. I'm Jacob Goldstein. Thanks for listening. Oh, also, Tom Hanks has something to say.

TOM HANKS: This is Tom Hanks. Do you know an undiscovered musician who deserves a break? Well, we have an idea for them. NPR Music is holding a Tiny Desk contest to find one great, unsigned musician to play the iconic Tiny Desk concert series and tour the United States with NPR Music. All you have to do is shoot a video of your musical act playing an original song behind a desk and submit it by January 29. Learn more at npr.org/tinydeskcontest.

GOLDSTEIN: One last bonus fact - so Ed Thorp - he came up with that formula to value call options. And remember, a couple of years later, some economists came up with the formula again, wound up winning the Nobel Prize. Well, one of those economists, Myron Scholes - he went on to be a key player at a hedge fund called Long-Term Capital Management. Long-Term Capital Management is famous today because it blew up and almost took down the economy, had to be bailed out. Thorp says they used too much borrowed money, too much leverage.

Copyright © 2017 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.