Michael Lewis's 2003 bestseller Moneyball exposed the public to a trend in baseball few casual fans had noticed: numbers-crunching analysts were taking jobs in major-league front offices and relying on the so-called sabermetric movementwhich evaluates players using statistics far more complex than batting average and runs batted into identify ways to gain advantages over other teams. Lewis described, in rather dramatic fashion, how Billy Beane and the Oakland A's built a perennial winner on a shoestring budget.
"In-game decisions require good old-fashioned horse sense and educated guesswork. The notion that they can be decided through statistics often is an illusion."
Shortly after the book came out, the A's began to falter. After averaging over 100 wins a season from 2001 to 2003, they averaged 91 wins a season from 2004 to 2006 and only 77 wins from 2007 to 2010. In their new book, The Beauty of Short Hops: How Chance and Circumstance Confound the Moneyball Approach to Baseball, brothers Sheldon Hirsch and Alan Hirsch, argue that Beane and modern front-office sabermetricians have come to harm the game: Though sabermetric thinking has helped advance the sport in certain ways, the numbers can obscure the beauty of the sport.In the following interview, edited for length and clarity, co-author Alan Hirsch discusses Moneyball, its fallout, the pratfalls of the modern sabermetric movement, and what he finds most beautiful about baseball.
Gelf Magazine: How do you define sabermetrics? When you refer to "sabermetricians" in The Beauty of Short Hops, whom are you referring to?
Alan Hirsch: Sabermetrics is the application of advanced statistics to baseball. Except for Bill James, sabermetricians are not household names. Many of them write for Baseball Prospectus. Others have landed front-office jobs with major-league teams.Gelf Magazine: That seems like a pretty important distinction, because I've seen the term sabermetrician used in a wide variety of contexts. In the book, are you talking about the people actually crunching the data, and not necessarily the fans reading Baseball Prospectus and Fangraphs? In other words, in your view does simply accessing that information obscure the beauty and intricacy of baseball?
Alan Hirsch: No. In fact, as you know from the book, my coauthor and I love baseball statisticsused properly and kept in perspective.
Gelf Magazine: On a conference call in November, Mets executive Paul DePodestaone of the heroes of Moneyballsaid, "Moneyball doesn't have anything to do with on-base percentage or statistics. It's a constant investigation of stagnant systems, to see if you can find value where it isn't readily apparent." Do you think it's possible that the book itself, by exposing the market inefficiencies that Billy Beane's front office capitalized on, forced Beane to search for different ways to win?
Alan Hirsch: Any sensible general manager will look for value that isn't readily apparent, and there are different ways of doing that. No small-budget team has succeeded more than the Minnesota Twins, an organization that openly disdains advanced statistics. But at the heart of Moneyball is the argument that Beane exploited market inefficiencies through his reliance on advanced statistics. As we argue in Short Hops, that's a distorted explanation of his success. Regardless, Beane is shrewd and will look for different ways to win. Indeed, he has recanted some of the central ideas that, according to Moneyball, accounted for his success.
Gelf Magazine: At the risk of getting into semantics, do you think Beane is necessarily recanting and not merely revising? Isn't it the nature of his job to be constantly adjusting, trying to find and maintain some advantage over his colleagues?
Alan Hirsch: You're absolutely right that a general manager should maintain an open mind. But one of the questions we ask in Short Hops is whether the alleged insights Moneyball touted involved a superior understanding of baseball or just new ideas that came into vogue and would eventually fall out of favor. Moneyball makes a big deal about Beane not drafting high-school players and thinking that baserunning is overrated. He is now drafting high-school players and the A's were among the league leaders in stolen bases last year.Gelf Magazine: As for the Twins, you mention that they openly disdain advanced statistics, and a quote from Ron Gardenhire in the book supports that point. ("I like the human element and I like the heart way better than I like their numbers. And that's what I'll always stay with.") But you do note that they use OPS and WHIPhardly wOBA or SIERA, granted, but stats nonetheless. Obviously all teams use stats to some extent to measure players. At what point does it become too much?
Alan Hirsch: That's a great question. One answer is that it becomes too much when the "knowledge" that's yielded is illusory. Take Ultimate Zone Rating, the new defensive measurement of choice. It made a splash when it showed that Derek Jeter was overrated in the field because his range was weak. Then, in 2009, Jeter's UZR soared. Are we to believe that, at the age of 35, Jeter's range dramatically improved? In fact, as we show in the book, UZR is a deeply flawed measurement.
Other measurements are simply too complex or arbitrary to be of much value. Take "linear weights," a formula that was supposed to capture a player's offensive value: Runs = (.46)1B + (80)2B + 1.02 (3B) + (1.40)HR + (.33)(BB + HB) + (.30)SB (.60)CS (.25) (AB H) - .50(OOB). I'd say the Minnesota Twins are wise not to pay attention to that sort of thing.
Gelf Magazine: Undoubtedly the formula for linear weights appears overwhelming, but in concept assigning run values to individual outcomes seems to make sense. And if teams are inevitably going to use stats to compare players anyway, shouldn't they strive to find the ones that best assess players' values?
Alan Hirsch: Of course you should look for the best stats possible, as long as they're usable. We certainly don't criticize teams for using statistical analysis as best they can. What we argue, though, is that often the value of the analysis is far more limited than people realize.
For example, we discuss Mickey Mantle's and Roger Maris's dream season, 1961. Who was the more valuable player? You can crunch the numbers every which way, using linear weights or win shares or any other formula, but there's one crucial number you'll probably miss: zero. That's the number of intentional walks Maris received in his record-breaking season. Why? Because of Mantle batting behind him. Mantle may have been of tremendous value to the Yankees just by kneeling in the on-deck circle, but that sort of thing can't be quantified.
We also talk about how statistical analysis offers only limited help for making in-game decisions. We give a number of examples drawn from baseball's rich history, such as Babe Ruth caught stealing to end the 1926 World Series and Grady Little leaving in Pedro Martinez in Game 7 of the 2003 ACLS. Analysts have reached tidy conclusions about these cases using statistical studies, but the reality tends to be much messier. In-game decisions require good old-fashioned horse sense and educated guesswork. The notion that they can be decided through statistics often is an illusion.
Gelf Magazine: The strength of the book, I think, is the vivid descriptions of baseball's oddities in Chapter 5 and the near day-by-day retelling of all the strange things that happen in a single MLB season (2009) in Chapter 6. Why are rare or random events at odds with advanced statistics?
Alan Hirsch: I appreciate your calling attention to that part of the book. It's something that everyone should get a kick out of, from the diehard to the casual fan.
I don't think that rare or random events are necessarily at odds with advanced statistics. It's a question of emphasis. In the book we quote someone saying, "While I still believed that numbers could reveal things about the game that were invisible to the naked eye, my own eyes had glazed over as the combination of fantasy baseball and mathematical arcana conspired to squeeze the life from the game I loved." To me, a big part of the life of the game lies in those crazy quirks that can't be captured or predicted by statistics. By the way, that quote isn't from some old-school guy who hates statistics. That was from John Thorn, one of the guys who created linear weights. He later saw the light.
Bill James has kind of seen the light, too. He's written that all that advanced statistical study has done is "to take a few buckets of water out of the ocean of ignorance and move them into the small pond of real knowledge. In reality, the ocean of ignorance is larger than it ever was, as it expands on its own." It's funny, but in some ways Short Hops is an elaboration of themes that James has embraced. I'm sure he'd love Chapters 5 and 6!
Gelf Magazine: Can I take it from both your response to the favorite teams question before and from Chapter six of Short Hops that you're a fan of the Red Sox, a team that has fared pretty well under a front office inclined toward advanced statistics (albeit one also blessed with a huge payroll)? How did you and your co-author come to the concept for the book? In other words, if your appreciation for the game's beauty is unaffected by sabermetrics, why peer into the sausage factory?
Alan Hirsch: To point to Oakland's division titles a decade ago, or Boston's championships, and declare sabermetrics a success, is rather selective. Why not also look at the other half-dozen saber-friendly teams, some of which have done poorly? Why not consider whether Oakland was able to sustain its success? Why not talk about the Twins, who succeeded while disavowing advanced statistics? Shouldn't we look at the whole picture? Contrary to what some critics (who admit having not read the book!) assume, we credit Bill James with rescuing baseball from a tradition of ignorance and discuss how he and his followers have advanced baseball understanding. And we certainly give Billy Beane credit for his success on a shoestring budget at Oakland. But why should Moneyball and the revolution it announced and accelerated be beyond scrutiny or above criticism? I can't see why anyone would object to a book which evaluates a movement that has come to permeate baseball.
As for our appreciation of the game's beauty, that's what the second half of the book conveys. The book became a workable concept in my mind precisely when I saw how the two partsthe critique of advanced statistics and the chapters about baseball's quirky spontaneityfit together. Ours is a vision of the game that gets less emphasis, that moves fewer people, because of an increasing obsession with numbers. But anyone who wants to keep making sausages will and should. That doesn't mean their work shouldn't be scrutinized by someone outside their circle. The irony here is that when Moneyball was reflexively criticized by old-school folks, those who liked it rightly rejected this knee-jerk response. Is a reflexive defense of the new status quo any more justified?