While author Michael Lewis was researching his book “Moneyball,” he was in the Oakland A’s locker room and noticed how unfit the naked players were when coming from the shower. When he mentioned to the manger that no one would know these guys were baseball players, the manager responded, “That’s the point, because those players were undervalued in the marketplace and one part of that was the way they looked.”
Much like an investment manager, the Oakland A’s were going into the market looking for undervalued products and “inefficiencies.” Lewis spoke to a packed room of investment managers attending the Morningstar Investment Conference in Chicago on Thursday.
Lewis said the A’s reduced risk by eliminating players who only played a few games, thus “small sample sizes,” and those who were very good and flashy at a particular skill. But those with subtle skills tended to be undervalued and were of interest to the A’s. In other words, the team dissected statistics and analyzed data to find the best undervalued team. This became known as the “Moneyball” method and is now used throughout sports. In fact, when Lewis spoke with John Henry, owner of the Boston Red Sox, who was going to implement the Oakland A’s system, he asked Lewis how much he could pay him to not publish the book. Henry, having made his money in the futures markets, understood once a system is known, market inefficiencies are harder to find.
After “Moneyball” was published, one review pointed out that Lewis missed the point, which really was the basis of research studies in the 1970s by two Israeli psychologists, Daniel Kahneman and Amos Tversky, that looked at “undoing our assumptions about the decision-making process.” In fact, their work goes far beyond sports; it was used in areas such as behavioral economics, computer science, medicine and government. The result of this research was his book, “The Undoing Project,” which went inside Kahneman and Tversky’s studies and the relationship between the two men.
Lewis said their studies are still used by the Israeli army on determining who would make the best army officers. Typically, like scouts in baseball, the Israelis would use gut judgment, which often proved wrong. Using the Kahneman/Tversky studies, they were able implement a system that helped selection via data analysis. It became so successful, even the Pentagon called the Israelis in to discuss their methods, Lewis said.
“The point is experts make mistakes,” and analyzing data can help prevent gut calls that become bad decisions,” Lewis said. Even in the markets, the move toward indexing and passive investing took some judgment out of decision. “So there was a shift away from the intuitive judgment of experts and toward a more complicated decision-making process.”
But there are limits to the method, Lewis said. “If you want to find the leading edge of algorithmic decision making, sports is really it,” he said. Key is the input to the algorithm, but “you never know what you don’t know,” Lewis cautioned.
That said, the NBA’s Houston Rockets adopted the “Moneyball” system but also continued to use a personal interview process that though racked with problems, still could smooth out inconsistencies highlighted in the data. For example, one time a manager asked a potential player why he had such a great senior year in college; Did he quit smoking pot as he had been caught for in earlier years? The kid replied, no, the school just quit drug testing.