Michael Lewis discusses his new book, ‘The Undoing Project: The Friendship That Changed Our Minds‘
The Huffington Post calls Michael Lewis “one of the premier chroniclers of our age.” It’s hard to argue with that description. His best-selling books, including Moneyball, The Blind Side and The Big Short, have all been turned into blockbuster films. In his latest book, Lewis tells the story of two cognitive psychologists whose work more than a half-century ago on bias and critical thinking has left an indelible mark on a variety of disciplines. The Undoing Project: The Friendship That Changed Our Minds explores the fascinating relationship between Amos Tversky and Daniel Kahneman that helped produce seminal thinking about how the mind works – and how it can reach more accurate analyses. Lewis recently appeared on the Knowledge@Wharton show on Wharton Business Radio on SiriusXM channel 111 to discuss the book and explain how his latest project is something of a prequel to the data analytics strategies found in Moneyball.
The Undoing Project: A Friendship That Changed Our Minds by Michael Lewis
Knowledge@Wharton: This book ties back to Moneyball. Tell us how it came about.
Michael Lewis: Moneyball led to this book in an odd way. Moneyball is, in my mind, mainly about the way markets misvalue people. It happens to be baseball players, but it’s a story about this team, the Oakland A’s, that has fewer resources than the competition. They have to find different ways, better ways to find baseball players. They discover in the process that the market for baseball players is not efficient. There are good baseball players who aren’t appreciated, and there are not so good baseball players who are over-valued.
When they were doing their business at Oakland, they were aware that baseball scouts made systematic kinds of mistakes, and they were in the business of exploiting those mistakes. When Moneyball came out, there was a review by an economist named Richard Thaler and a lawyer named Cass Sunstein. They said, “Michael Lewis told a nice story, but he doesn’t seem to understand his own book.” They said that these biases that are in the mind of the scouts are cognitive biases, and they were described and uncovered by two Israeli psychologists named Amos Tversky and Danny Kahneman. I had never heard of these guys. I thought, “My God, how’d I miss that?” I never asked the next question, why? What’s going on in the mind? It hadn’t even occurred to me that anybody ever had.
It took eight years to do this book. I took a few years after that review to finally have the wits to call Danny Kahneman and say, “I want to just talk to you about this.” It turned out that he lived up the hill from me in Berkeley in the summers. So, I went up the hill and we had coffee.
All of the sudden we’re talking long walks in the hills, and I start to hear this story of his relationship with Tversky, and I realize that Moneyball and the whole phenomenon was one offshoot of their work, but that it crept into behavioral economics. It created behavioral economics. You found it had an influence in medicine and law and so on. I just thought the relationship, the characters were unbelievable, the relationship was this passionate kind of love affair without sex, but otherwise they were really kind of crazy about each other. And there was a great deal of drama in the relationship. It was an incredibly important scientific collaboration. Eventually, I got around to realizing that this is a book on its own. It’s a kind of prequel to Moneyball.
“The intuitive judgment of experts had this inherent fallibility in it.”
Knowledge@Wharton: Their research and thinking happened 50 years ago, correct?
Michael Lewis: That’s about right. They collide in 1969 at the Hebrew University in Jerusalem, and the relationship starts to bust up in 1979, 1980, 1981, by 1981-1982. By then, they’re at Stanford and Princeton. Amos Tversky ends up at Stanford. Danny Kahneman eventually ends up at Princeton.
So yes, it’s work that was done a long time ago. It was all worked out in kind of arcane, boring psychology journals. The work itself was not boring, but the context was so tedious. It wasn’t really aimed at anybody outside of psychology except for one article they wrote for a general interest audience in Science magazine in 1974. That reached people in all kinds of different disciplines. What happened was they infected people’s minds, and this took a while. Sometimes ideas take a while to take hold.
They explained to a lot of people why expert judgment was flawed, why you had to be careful about it. The intuitive judgment of experts had this inherent fallibility in it. Then along comes the information revolution, the computer revolution, and it becomes cheaper and cheaper to generate both data and algorithms to make these analyses and decisions that were previously done by people. That’s what the Oakland A’s would do, right? They’re trying to find new and better ways to collect and grind baseball performance statistics. Those calculations end up being the basis of their investment decisions, as opposed to asking some scout, “Is he any good?”
Knowledge@Wharton: Part of that was because the Oakland A’s didn’t have the big money to be able to invest. They had to figure out the best value for every penny in that team. That philosophy is now being used by so many baseball teams, even the ones that have millions of dollars to throw around.
Michael Lewis: There’s no reason to make stupid investments just because you have a lot of money. That’s what the Yankees and the Red Sox and the Chicago Cubs eventually figured out, and they hired kind of disciples of Billy Beane and people who were following the methods of the Oakland A’s front office. As a result, the Oakland A’s are in kind of a bad position again because the rich teams now have not only the money, but they also have the same kind of intellectual property.
Knowledge@Wharton: You call Daniel Kahneman and Amos Tversky two of the most amazing characters you’ve ever been involved with. Why so?
Michael Lewis: Everything that came out of their mouths and their minds was interesting. I knew that if I just got their words and their behavior down on the page, their characters would infect people’s minds. That people would start to think, what would Amos Tversky say, or what would Danny Kahneman say?
The kind of things people said about them was extraordinary. There’s a psychologist at the University of Michigan named Dick Nisbett who, after he’d spent a lot of time with Amos, designed a one-line intelligence test. The test was: After you’ve met Amos, the longer it takes you to figure out that Amos is smarter than you, the stupider you are. Everybody who knew Amos said, “Yeah, that’s right.” He wasn’t obnoxious about it. It was just he was kind of annoying. He was actually a Spartan warrior in the Israeli Army. He was a highly decorated war hero. He was playful and fun and all the rest, but he took what you said and made it so much more interesting than what you had said.
There were hundreds of anecdotes like this. He was at a party with some of the greatest physicists in the world, and they didn’t know who he was. He was there kind of by accident. And after the party, one of the young physicists who’d just won a very distinguished physics prize called up the hostess and said, “Who was that physicist I was talking to?” And he described Amos. She said that he wasn’t a physicist, “He’s a psychologist.” And the guy said, “That’s impossible, because he’s the smartest physicist I’ve ever met.” This kind of thing happened all the time. He had this kind of logician’s diamond-cutter mind. He just had an ability to see things that other people didn’t.
Danny was this font of creativity. Danny was kind of a poet/novelist type, although he insisted on being a scientist, and he just had the most startling original insights and ideas. To whom does it even occur that the imagination might obey rules, rather than just be this free-flowing thing, and that you actually might be able to study and classify the rules of the human imagination? And then design tests to do so? That’s the kind of thing he’d do in an afternoon.
“There’s no reason to make stupid investments just because you have a lot of money.”
I’ll give you another example. This is breathtakingly original and profound. Both of these guys were very heavily involved with the real world. They weren’t just sitting in a classroom. Because they’re Israeli, they’re off every six years shooting and getting shot at. But they were also training Air Force pilots and training tank commanders and so on.
Danny was training Air Force instructors one day, and he noticed they kept saying, “When you’re teaching a pilot, praise doesn’t work, only criticism.” He said, “Why is that?” And they said, “Well, when they do something great and we praise them, they’re always kind of worse the next time. And when they do something bad and we lay into them, they get better.” Danny said, “That’s called regression to the mean. What you’re seeing is an illusion. They did something great, it was a little better than what they normally do, but they have a mean level performance. The likely thing that’s going to happen next is they’re going to revert to the mean.”
He not only sees that but realizes that as teachers we’re kind of doomed to a lifetime of getting this reinforcement that our criticism works more than our praise, but just simply because that’s what’s going to happen when people are doing unusually bad or unusually good things.
This came out of his mouth when I was talking to him four or five years ago. I was coaching all my kids’ teams — I still do. And I tell you, it changed the way I coached. I had been noticing this. It seemed to kind of work when you criticize them. If they do something really bad and you criticize them, they did get better. And if they did something great and you praise them, it doesn’t seem to do any good. But in fact it’s baloney. It’s an illusion. So, I’ve made a huge point in my coaching that it’s 3 to 1 praise to criticism just to offset this tendency.
Danny Kahneman does not know a basketball from a football. He has absolutely no interest in sports. He thought it was curious that Moneyball kind of came out of his work in a funny way. But things come out of his mouth that actually shift your view of how to do something that he has no idea that’s even being done.
Knowledge@Wharton: Amos Tversky passed away several years ago. How much information about him did you have to gather from Danny and other people?
Michael Lewis: Amos died in 1996. I thought before I started writing that bringing Amos back to life was going to be the big challenge of the book. There were big challenges in the book, but that was not the challenge. Amos is so vivid. He was such a strong personality. He had a trait that was very useful in the biographer of a dead person, that he never did anything he did not want to do, in the most extreme way.
If he went to a movie with his wife, he would judge it in five minutes. If he sensed it was trite, he’d get in the car, go home, watch Hill Street Blues on the sofa and go pick her up at the end of the movie. He’d say, “They took my money. What, they want my time, too?”
Graduate students had endless stories about the huge piles of mail arriving for Amos. Everybody wants Amos. Everybody wanted to know what he thought. He would go through the mail just looking at the envelopes, and they’d go unopened into the garbage can, boom, boom, boom, boom. He’d open one or two. He’d look up and say, “I have a what-can-they-do-to-me rule. If they can’t do anything to me, I’m not going to open it.” He told people they wasted so much time worrying about social embarrassment and convention. If you were in a faculty meeting or at a party and you just feel like you don’t want to be there, don’t sit there worrying about what people are thinking if you leave. Don’t worry about what excuse you make up. If you just get up and walk out, your mind will form the words that you need to explain why you’re leaving.
Obviously, he offended some people. But he was so charming about it, and everybody kind of knew. As a result, everything he saved and everything he did, all the pieces of paper in his filing cabinets, every interaction he had with people, all the friendships are meaningful. He wasn’t doing anything just going through the motions. What he left behind says so much about him. It wasn’t that hard to re-create him. He’s such a great character.
Knowledge@Wharton: In an interview you did recently, you talked about their relationship almost like The Odd Couple. They were like Felix Unger and Oscar Madison.
Michael Lewis: Sometimes I thought, I’m writing about Felix and Oscar. Sometimes I thought I was writing Brokeback Mountain, but they screw each other’s ideas. When they traveled, they were mistaken for a gay couple because they were so close. But they were completely straight. But they were more attached to each other than they were to anybody else in their lives. But the Felix and Oscar thing, absolutely. Even they were like Felix and Oscar in the way Amos was a neatnik. People said you go into Amos’ office and it was the most extreme thing. There was a pencil lined up in the middle of the desk and that was it. If Amos was working, he reached into the drawer next to him and pulled out of a pad of paper, and it was square to the desk. Then he’d put it away when he was done. There were no books on the shelves, no paintings on the walls. It was just spare.
Danny’s office was such a catastrophe that his secretary tied his scissors to his chair so she didn’t have to look for them. People said you couldn’t find anything in Danny’s office because it was chaos, and you couldn’t find anything in Amos’ office because there was nothing there.
“What they identified were essentially the tricks the mind plays on itself in various circumstances.”
They were opposite types. Amos was the life of the party everywhere he went. I heard some version of this description over and over. Amos walked into the room, and nobody really noticed him. He wasn’t particularly physically distinguished in any way. Kind of small, unassuming, and he didn’t dress in any particularly unusual way. He’d sit and listen for five or 10 minutes, then he’d say something and everybody would turn to him. They’d say 20 minutes after he arrived, everybody was around him like moths to a flame. He would just mesmerize people, take over the room.
Danny was distant and removed from people, very formal. Danny was, at heart, kind of a French intellectual. Aloof is the way it read. However, Danny is the one who wants to mix it up with the tanks drivers and the Israeli Air Force pilots. He did Moneyball for the Israeli Army. He completely transformed their selection system in a way that the Pentagon’s calling up and asking, “What the hell are you guys doing because something’s working.” He wrote an algorithm to choose officers that they still use today. He did it in 1954 when he was 22 years old. It’s incredible. I went with him to visit the Israeli Army base, and he’s worshiped like a god. They call the score they give young kids to determine where they go in the Army the Kahneman score.
Amos was this very detached intellectual. If he had never met Danny, you’d never have heard of him because all he would have been doing is math formula. He thought of himself as a mathematical psychologist, and you don’t want to know what that is. He did not naturally engage intellectually with the rest of the world until he met Danny.
Knowledge@Wharton: Does it surprise you that so much of their work still plays out today in so many different areas? You mentioned baseball, but it could be very easily health care or the presidential election.
Michael Lewis: What they identified were essentially the tricks the mind plays on itself in various circumstances. They were showing that although life constantly puts you in these probabilistic situations, these situations that might lend themselves to statistical analysis, and we don’t do that. People aren’t natural statisticians. They do something else. What they do is tell stories. They find patterns. Danny and Amos were showing the way the mind, when it’s telling stories to resolve uncertainty, makes mistakes. It does apply to everything. Everything goes through the mind.
“There is no such thing as a great quarterback. It’s a great quarterback in the right situation.”
The question is why, having identified these cognitive illusions or whatever you want to call them, they persist. We don’t pay more attention to them. I think they would say that what we found in the mind was that the eye tricks you, the ear tricks you. We can show you even how it will trick you. There are a lot of really reliable optical illusions. Even if I tell you that that’s not water on the desert highway, that’s a mirage, you know intellectually, yeah, that’s right. You still see the mirage. The optical illusion doesn’t go away.
You could say the same is true of the cognitive illusion. It doesn’t go away. It’s very hard for a person to self correct. What you can do, Amos would say, is change your environment in which you make decisions, so people are more likely to point out to you if you’re making errors. It argues for decision-making environments that aren’t autocratic and an approach to decision making where the decision maker isn’t assuming he’s infallible or has unbelievable gut instinct.
You build checks into the process. One kind of check you can build into the process, which is what the Oakland A’s did, is have good data about the decision you’re making. I’m looking at a baseball player. He looks like he could be a great major league baseball player, but it’d be nice to have performance statistics that give me the sense of the probabilities of him being a good baseball player.
It’s not that there is an algorithm that is going to give you the perfect answer. But it is that data can help you make the bet, queer the odds in your favor a bit more. And the eye does really play tricks on you.
Even though it’s about these two Israeli psychologists, the book opens with a long section inside the mind in the front office of the Houston Rockets, because they are on the edge of trying to rely exclusively on data and analytics in their decision making. Even they see that eventually it hits its limits because instead of being wrong 60% of the time, you’re wrong 50% of the time and everybody’s still a little disgruntled. So, you let the expertise, the intuitive judgment back into the process. Now you’re dealing all over again with the cognitive illusion.
Knowledge@Wharton: How much do you think that environment plays a role in that? There have been baseball players that we thought were going to be so successful in the minor leagues, but it didn’t work out when they went to the major leagues because of the situation they ended up in.
Michael Lewis: This is even more true in other sports. When you’re playing baseball on the Yankees, versus the A’s, versus a minor league team, it’s largely an individual sport. But if you are a football quarterback, you might be a superstar in one system and actually wash out in another system. If someone had asked Peyton Manning to run a wishbone offense or be a running quarterback, that wouldn’t work.
It’s absolutely true that there is no such thing as a great quarterback. It’s a great quarterback in the right situation. The same is true to a lesser extent the other positions, and it’s certainly true in basketball. Part of the problem Jeremy Lin had post Knicks was he was in exactly the right situation in Madison Square Garden for a little bit. He really is at his best when he gets to have the ball in his hands. He gets to the Rockets, and he’s not quite as good as James Harden in that role. He’s put into a different role and it’s just uncomfortable. But this is a different subject. It’s a fun subject, but a different subject.
Knowledge@Wharton: The fact that Dan Kahneman has that much influence in the Israeli Army 60 years later is staggering.
Michael Lewis: You can trace their influence onto Wall Street. In the beginning of the move away from managed stock picking to index funds, their writing is there. Their work is right there. You can trace it into medicine where doctors start to question the wisdom of diagnosis of disease. Amos comes along, and they do work where they show that if I tell you there’s this operation that could cure your cancer, but there’s a 10% chance you’re going to die on the table, you’re far less likely to have the operation than if I tell you there’s a 90% chance you’re going to survive. I mean, that’s crazy. One of their insights was that people don’t decide between things, they decide between descriptions of things.
Article by Knowledge@Wharton