Moneyball

"He reads a book called Moneyball — the Michael Lewis book about the Oakland A's — in almost what seems like an instant decided I really need to be in sports analytics."

"Yes, this is a book by Michael Lewis. I’m a great admirer of Michael Lewis . He’s by far the most accomplished, most entertaining, most informative non-fiction writer around at the moment, and I’m slightly surprised he’s not better known in Britain. I wanted to include this book because it seems to me a parable of the disinclination of people to base their practices on evidence, and for evidence-based policy in general. The book focuses on Billy Beane, an interesting figure who is the general manager of the Oakland Athletics. The Athletics punched way above their weight in the time leading up to the writing of the book, and Lewis got interested. How did they manage to hold their own when they had a third of the money to spend that the Yankees and the other big teams had? As he tells the story, Beane was the first Major League Baseball figure to pay any attention to something that had been going on for about 40 years before this, which was geeks analysing baseball statistics and figuring out that the supposedly tried and tested strategies of the managers, who were mostly old baseball players, were completely misguided, and that you could clean up just by making a few simple strategic changes. This whole movement is known as ‘sabermetrics’ (from the acronym SABR from the Society of American Baseball Records) and was started by an eccentric statistical analyst and writer, Bill James. He got the movement going in the late 1970s, though it goes back even earlier. I don’t know how many people have read Philip Roth’s book , The Great American Novel . It’s difficult for English people to read because it’s a fantasy, picaresque, non-realistic, extravaganza about an imaginary baseball team in the 1940s, and it’s Roth at his most playful and untrammelled. It’s a lot of fun. One of the baseball teams in that book has an owner whose son Isaac Ellis is a self-styled ‘17 year-old Jewish genius’, and he goes to all the players and the manager of his father’s baseball team, and says, ‘Look, you’re doing it all wrong, you really shouldn’t be making sacrifice bunts, that’s costing us 70 runs a year, and all this stealing bases, it’s even worse,’ and he demonstrates it all with graphs, and they all say, ‘You’re nuts, please go away.’ This book was written in 1973, and Roth credits Earnshaw Cook, who’d written a book developing these ideas back in the 1960s, so the facts were out there for 40 years before anybody in baseball took any notice. This is odd because it’s a competitive sport, you want to win, and there are huge amounts of money involved. Nowadays the players are paid ten million dollars a year salary, and even back in the 1970s, ‘80s, and ‘90s, there were significant amounts of money involved. Yet nobody ever bothered to do the sums to figure out whether what they were doing was sensible. That’s part of it. Billy Beane, the central figure, was himself a great young baseball hopeful, he looked completely wonderful, and he played in Major Leagues on and off for a few years, but he never batted well enough for the level at which we was playing. That sort of weakness does get found out, it’s like a plausible-looking striker who never scores any goals. You’d see at the end of the season, he’d only got five goals, not 20. Billy Beane was averaging in the low 200s as a batter, and that’s not good enough. So it’s not as if people who look good but don’t perform don’t get found out: it’s more subtle. Baseball batting averages are the traditional way of assessing the worth of a batsman. But the statisticians showed that batting averages are a terribly crude tool, and RBI (runs batted in), another statistic they use all the time, is another terribly crude tool: they don’t measure what you’re really interested in, they don’t measure how much the batters are contributing to the success of the team. If you’re no good, you’ll be found out even by these crude measures, but if you really want to work out which are the best players to spend your money on, which tactics you should be encouraging, then you need a very careful statistical analysis, and nowadays, all the baseball teams do use it. The shocking thing is that the data and information were available for decades, and somehow they just all ‘knew better’. They knew better in their guts that the right way to play was the way they’d always played, and it makes one think that, well, baseball managers, they are a kind of in-between case: they’re baseball managers, they spend all their lives in baseball, they have all these very strong traditions and ideas, so that’s going to count against them innovating. On the other hand, they are in a very competitive environment involving lots of money, and you think that analysts would come in and tell them what to do, but they didn’t. The analogy that one immediately wants to draw, and of course it’s worrying, is politics in general. There are politicians and people who vote for them who think that they know in their guts what’s right about immigration, drug policy, prison policy, economics in general, and they will vote and the politicians will follow them just on these gut reactions; even when there’s plenty of statistical evidence around showing that the policies everybody favours are not going to produce the effects they want. I think the lesson’s probably been learnt within sport. Famously in baseball, the Boston Red Sox, who never won the World Series since they sold Babe Ruth back in 1920, tried to hire Billy Beane, and he wouldn’t go. Then they hired one of his disciples and they won the World Series a couple of years later. So baseball has learnt its lesson. My understanding is that soccer now uses very similar techniques. When there are huge amounts of money involved and plenty of people coming out of business schools all ready to help—sports enthusiasts themselves—it would be surprising if one didn’t see this kind of analysis informing decisions in all kinds of sports; it’s just a pity it isn’t used more generally outside sport."

"I discovered it when I was in the US and was a visiting writer at Harvard in 2003. I think I was one of the only English writers to talk about this book when I came back to Britain at the end of that year. I wrote a review of it and I tried to use it practically when I was captain of Middlesex – we used some Moneyball -like statistical methods to inform our selection of our Twenty20 teams. So my interest in it started off as intellectual, but then became practical. Of course, the idea of using statistics to gain an edge is anti -luck if you like; it’s about using numbers to gain an advantage. But it’s more complicated than that I think. Moneyball is a superb book and credit should go to Michael Lewis for turning a book about baseball statistics into a bestseller and for attracting readers who do not like or have any interest in baseball. He makes the strongest possible case for using thinking to help win more games and using statistics to work out how a baseball match is won. And it’s not won in the way that we all think it’s won – by just scoring runs in the obvious sense. He breaks down how those runs are scored, who it is who is scoring, in much the same way as a financial analyst would do it. Since then, every baseball team, including the very richest, have tried to adopt this method. For some it has worked, for others it hasn’t. What I take from the book now is that we would be crazy not to use the latest mathematical tools to help understand how games are won and lost, but I think the book takes to an extreme the idea that human judgements are absolutely obsolete in the face of scientific data. That idea has been shown to be a little dated. It’s almost as if Moneyball was the high watermark of enlightenment self-confidence – the idea that the numbers will tell the truth and human beings, with their gut instincts and judgements, will inevitably make mistakes. “We would be crazy not to use the latest mathematical tools to help understand how games are won and lost.” A lot of baseball teams haven’t benefited from the Moneyball methods. The Oakland Athletics team did, but that was assisted by their business model. I think success in sport will always be a mixture of statistical based analysis and human judgement. One without the other will never be enough. I think what Michael Lewis did brilliantly was expose the clubbability of professional sport. There is a tendency to be very suspicious of new thinking and I experienced this too. So he’s saying: “Hang on a minute, these guys are just recycling conventional wisdom and are closing themselves off to new and better ways of doing this.” I completely agree with him on that. But what I wouldn’t agree with him about is the idea that a computer can make all the right decisions about selecting players. That does not fit with my experience nor indeed of the evidence of today’s sport. So I don’t think that statistical modelling, whether in sport or financial risk, will ever supplant a shrewd combination of modelling and human judgement."

"Everybody knows this now because Brad Pitt is in the movie. It has to be presented in a different way, but it is a good reflection of the book, yes. It hits all the high points as well as being wonderfully entertaining. For example, in both the book and the movie, you really do get a feel for this guy, Billy Beane, who everybody said looked like a great baseball player. He wasn’t, for a remarkable variety of reasons, none of which you can put your finger on. He got the idea that if you started thinking about this (i.e. how you recruit new players to your team) in a formal way, you can do better than people who just do it based on their gut. I’ve assigned this book to a ‘first week of the year’ group a couple of times. The guys think it’s wonderful, though sadly the women did not find it very interesting. I also have kids read it in my econometrics class. There’s a huge amount of econometric research on baseball . What the book is really about is a guy who is using some econometrics to predict which players will do better in advancing wins, which is a remarkable use of economic thinking. These guys really thought out of the box. Yes, he’d done some modelling. But without Billy Beane’s feel for the game, and willingness to use it, it wouldn’t have happened either. Like most good things, it’s rarely just one person – it’s a juxtaposition of a couple of people’s views that makes a nice innovation. There was a paper written by Raymond Sauer of Clemson University , published four years ago. The key point of the paper, and the point of the movie, is that the big guys, like the New York Yankees (whom I hate with a passion and have since I was five years old) have all this money. Of course they do well, because they can buy the best. Nonetheless, for a few years, because of what Billy Beane and his statistical people did, the Oakland As did phenomenally well on a truly miserable payroll. What was neat, that this paper showed, is that within three years, the As were back on the regression line. In other words, the innovation that they created spread very rapidly, so everybody started doing the same thing, and their competitive advantage disappeared. It’s another great illustration of how competition can erode an innovating entrepreneur’s advantages. If you look at their ability to increase the chance of a win, people used to look at batting average, which is hits over at bats. But an at bat doesn’t occur if you get a walk, or if you do a sacrifice fly. But a walk is as good as a single. So what he started looking at is bases or on-base average, and using these to predict how well a player would do. If you do very well at getting on base, if you walk every single time, you’re golden, but your batting average would be zero. So he started thinking about different statistics and people started putting these inputs into a production function for runs and wins. That’s the technical innovation. Some people get walks a lot. For example, during the World Series, Albert Pujols walked about three times in a row because people didn’t want to pitch to him. They were afraid he’d get a hit. That didn’t help his average, but it made him worth a lot, and people pay off on that. He’s paid a bundle. Some people, for whatever reason – they have a stance, or because of their size, or the fear they instill – get walks. That’s just one example. But the book is neat because it brings together both this technical innovation and the very human story of this guy who was going to be a superstar, and didn’t quite make it. The movie doesn’t quite give you the same feel of sympathy for the Brad Pitt character as the book does. I found him more sympathetic in the book. In the movie there wasn’t sufficient time to develop him as a character. He’s quite a good-looking guy, actually. They can apply it to almost anything they do. That’s what’s so neat. This stuff is just everywhere, like dealing with your roommates. Your roommates’ relation with you is a game theoretic problem. Who does the dishes? Who does the cleaning? For anything involving interpersonal relationships, the stuff we do in economics will be useful. On the first day of class, I tell my students: “You think economics is about money. It’s not. It’s about scarce resources.” Even if you have all the money in the world. It so happens that one of the richest people in America lives in this town, Michael Dell. He’s tremendously constrained because he has only 24 hours, just like you and me. So much money to spend! So little time! So these are the ideas I try to get across. I’m one of the few remaining people who are regular professors, who teach large classes. Mostly it’s farmed out to temporary staff, sadly enough. I think this is a loss, because if we just tell our kids this is a bunch of formulas, or a bunch of graphs, and don’t give the appreciation that, firstly, it’s about the real world, secondly, that it’s a more useful way of looking at the real world than any other and thirdly, that it’s fun, we’re losing a whole lot of people who will never find out what economics has to offer."

"My choice, I can assure you, had nothing to do with Brad Pitt. I chose Moneyball because it’s a wonderful book and because it influenced my own. Moneyball was published right before I wrote Mindset and it showed that the fixed mindset was alive and well in the world of sports. You would think that the relationship between training and skill would be utterly obvious in sports, but apparently it isn’t. Many of the baseball scouts described in the book really thought they could look at superficial physical features of baseball players and know who had the potential to be a superstar. It’s the sports version of craniometry. The book is built around Billy Bean, who as a young man was identified by baseball scouts as the next megastar. However he was a dismal failure because of his fixed mindset. He thought everything should come naturally. He disdained practice and he got hung up about mistakes. Every time he struck out, which even the best players do most of the time, he had a temper tantrum. The book tells the story of how Billy Bean went from having a fixed mindset that doomed him as a baseball player, to having a growth mindset that made him one of the greatest general managers of all time. How do we go from a fixed mindset with all its fears to a growth mindset with all of its opportunities? One thing you can do is start listening to the voices in your head. Each mindset has a voice that says different things. With a fixed mindset, as you’re approaching something that’s really difficult the voice says: Watch out, maybe you’ll humiliate yourself, maybe you’ll show yourself that you’re not as smart as you think you are. But the growth mindset answers back: You’ve got to try this, you’ll never improve unless you try and everyone, after all, is a novice before they’re an expert. As someone is struggling with a task and making mistakes, a fixed mindset voice says: See, I told you. You’re making a fool of yourself. Look at these mistakes. Obviously you’re not good at it. If you were, you wouldn’t be struggling. The growth mindset replies: It’s called learning, and learning happens over time through persistence. If you want to hold your head up high you’d better keep at it. When someone with a fixed mindset sees someone who is really great at something, the voice says: That’s what talent looks like and you don’t have it. You’ll never be that person. In fact, studies show that people in a fixed mindset are not inspired by role models, they’re intimidated and demoralised by them. But the growth mindset voice must answer back: That’s what you could become. Learn more about how that person did it so you can do it too. Another important thing is to understand the neuroscience behind the growth mindset – how your brain changes with learning and how you can actually shape your brain through your thoughts and actions."

"Any baseball enthusiast will know that there’s a type of baseball fan who is very interested in statistics. If you follow players or teams, people are always quoting statistics, most notably the batting average, or how many home runs a person hits. There’s a statistic for just about everything you can do on the field. I’ve been a longtime baseball fan but I’ve not been a fan of the statistics, whereas a friend of mine whom I often go to baseball games with, he’s obsessed with the statistics and I’ve never understood why. Someone said to me, ‘you ought to read Moneyball by Michael Lewis, it’s about baseball statistics’. I was familiar with Lewis’s other books and I thought, baseball statistics? But what Lewis does is he mixes the statistics with the life story of a manager, Billy Beane, from his high school days all the way up. Beane turned out to be a baseball genius because he realised that some of the numbers people dwelled on were really not important for predicting outcomes of games. Other numbers were more important. Beane managed his players completely by statistics, and he took his low-budget team, the Oakland Athletics, to an unprecedented series of successes that won him wide acclaim throughout Major League Baseball. I like this book because statistics is a very dry subject , but Lewis makes the numbers come to life by integrating them into Beane’s life story. I thought I could use a similar approach in writing about radiation, because so much of radiation is statistics, particularly in estimating risk. It’s very dry and hard to understand, but I tried to use some of Lewis’s approaches to describe things in a way that I hoped would speak to people. Fallout is radioactivity that mixes with dust and then settles down on people. It’s the second consequence of nuclear weapons that is just as deadly as the blast radiation. But before the atomic bomb test in Bikini, we didn’t know much about the problem. That’s because at Hiroshima and Nagasaki there wasn’t much fallout; the bombs were detonated above ground so they didn’t kick up any significant amount of dust. When the tests were done at Bikini, the bombs were much larger, and they were detonated at ground level, so they kicked up a tremendous amount of fallout. “Fallout is radioactivity that mixes with dust and then settles down on people. It is just as deadly as the blast radiation” These fishermen were in the wrong place at the wrong time. They happened to be fishing downwind of these detonations. The fallout fell on them like snow, it was several centimetres thick on their boat. They didn’t know what it was all about. They returned home and all got radiation sickness, they were in hospital for a year and one of them died. This was the first time we really understood how bad fallout could be."