"we can only do what it seems to us we were made for, look at this world with a happy eye but from a sober perspective.”

-W.H. Auden

The Revolution in Stats

I am a nerd. This fact was quite apparent to many of those around me growing up, but came as quite a surprise to me. Part of the reason that I was regarded as a nerd was because I wasn’t into sports. Oh, my father, a native Chicagoan, enjoyed basketball, and I was at least aware enough to bask in the glow of the Michael Jordan years. But in most ways, I was an outlier from my baseball/basketball/football-loving peers. In fact, I'm sure I must be in a very tiny minority of healthy American males who have never played organized baseball. I was a three-sport athlete in high school—generally a poor one, though I had my moments—but not in team sports. It wasn't until my later teenage years when my appreciation for the great American team sports flowered.

Being a nerd, of course, I developed my love through study, through reading. I remember reading an entirely forgettable book about boxing in high school—so forgettable I couldn't possibly tell you the title—that nevertheless inspired me to explore the grammar and mythologies of sports, which had been so foreign to me for so long. Through reading, sports ceased to be a private vocabulary—one that every other boy seemed to have had whispered into his ear at infancy, but which had strangely been denied to me—and became instead a new intellectual problem, something else to be considered and solved.

The thrill of sports is and will always be largely visceral. I would have it no other way. But behind the moments of raw action are endless intricacies, seemingly limitless geometries of movement which can be studied and enjoyed in precisely the same way one enjoys science, math or history. I'm sure some people are probably reading that sentence in horror--the division between jock and nerd is so elementary and animalistic I'm surprised Joseph Campbell never wrote about it--but I mean merely that intellectual play in the consideration of sports is little different than in any other subject. There is something universal in the basic pleasure of applying mind to (subject) matter and slowly, gradually, feeling the unknown become the familiar.

My sports appreciation proceeded in the kind of idiosyncratic, organic way that is typical of most types of intellectual growth driven by discovery and joy. I had grown to love sports within the course of a few short months. But it wasn't until I played fantasy sports that I came to love stats.

In 1998 or 1999, I was fortunate enough to be invited into a fantasy football league that some friends of mine had started some six or seven years earlier. The league started, that is, in 1992, when I was 10 years old—ages before most people had heard of fantasy sports. Back then, our commissioner would add up the scores by rifling endlessly through newspaper sports sections; game results were dispensed in hand-written notes, distributed by hand in the cafe at school. (He was once approached by his boss at a summer job, who had observed his endless time spent combing through box scores, and asked very seriously if he had a gambling problem.) Fantasy football led to rotisserie baseball and fantasy basketball, to March Madness pools and fantasy golf, to leagues run entirely through computer simulation, and on and on.

In these leagues, stats were indispensible. Before, working with stats had often seemed like doing work, but now they granted insight, an edge. Stats began to seem like a secret code, a shared understanding—in exactly the way, I must admit, sports had long seemed like a private language. Only now, I was on the inside. And no sport seemed to present more new, little-known stats than baseball. Every new stat and metric represented a new way for me to understand the game in a way that, for a while, had not been appropriated by the general public. For a while, but not for long.

Sabermetrics, the application of new statistical and analytical schemes for understanding baseball, has in the span of a few short years gone from being the passion of a small, dedicated core to being as mainstream and all-American as Sports Illustrated. Still sometimes referred to, unhelpfully, as "Moneyball," after the book by Michael Lewis, sabermetrics began as a product of the Internet revolution. As has happened with many previously marginal intellectual movements, new statistical visions of baseball were incubated, then amplified, by the connective power of the Internet, where like-minded individuals could collaborate and the old guard could reach out to a new generation dissatisfied with the traditional tools for evaluating baseball. This change was met, quite notoriously, with no small amount of resistance from many within the sport and the media; baseball, more than any other American sport, is concerned with tradition, history, and the past. Today, however, sabermetrics counts among its converts not only many in the baseball media (in print, television and online) but also more and more front offices of MLB teams—previously the province of hidebound traditionalists. The value and prominence of sabermetrics has built to a kind of critical mass in the last year or so; just recently, the highest authority of the American sports fan middlebrow, Bill Simmons, gave sabermetrics his full endorsement.

All of this probably sounds quite triumphalist, and very pleasing to my fellow converts. And yet as sabermetrics becomes more and more mainstream and prevalent, I grow more and more troubled.

Across many disciplines and fields, you can observe consistent patterns coming from, shall we say, insurgent intellectual movements. In art, history, and science, change often proceeds from the repudiation of some cherished knowledges held by the old guard. The dynamic of fighting against entrenched, hidebound authorities who resist new ways of knowing has consequences for the upstarts, some good, some bad. Working from a position of enforced disadvantage, the intellectual insurgency must develop an argument that is concise, accurate and airtight.

Yet even when such developments are, as a whole, to the good, they almost always come at a price. Ruthlessness and efficiency are powerful forces when waging a battle, intellectual or otherwise, but they leave little room for the kind of probity, restraint and discretion that are at the heart of responsible intellectual discourse. Forgive me for waxing Hegelian, but the goal, to my mind, is not to banish older ways of knowing from the kingdom of the good, but to take what is useful in that knowledge and synthesize with new ways.

There are, after all, other ways of understanding baseball and its strategies, the kinds of knowledge that come from accumulated experience as a player or coach. Of course, playing or coaching is no excuse for ignoring or disparaging statistical evolution, and yet there have been some prominent ex-players and coaches who have done exactly that. I don't pretend that there isn't a reason why many supporters of new metrics have a certain suspicion towards analysts who played or coached the game, but to reject the legitimate knowledge held by these people does our project--better understanding the game of baseball—no favors.

To be fair, most of those performing the important work within sabermetrics are quite level-headed and respectful of other kinds of baseball knowledge—the kinds that have been developed by those who have played and coached in the game for decades. But those actually involved in the compilation and analysis of these stats tend to speak with far more restraint than some of their followers.

I am only beginning, as a graduate student, to become informed and credentialled in statistics. Even from that position of near-amateurism, however, I know that there is a danger in dividing statistical analysis from a larger statistical literacy, or from the philosophical framework that limits and qualifies statistical knowledge. Statistics is a cautious discipline that produces information that is often used recklessly. This is hardly a problem with sabermetrics alone; in our culture, we are constantly told that "studies say" or "research shows" when the actual language of those studies and that research is usually far more complicated and equivocal. I know that there are many working with modern statistical techniques in baseball who are perfectly aware of the standards of statistical accuracy and limitation—indeed, they are far more educated on that subject than I am. The experts who are actually involved in the compilation and analysis of these stats also tend to be level-headed and respectful of other kinds of baseball knowledge—the kinds that have been developed by those who have played and coached the game for decades. I only wish they were better at empirical modesty through their work, so that the rank and file of the sabermetrics crowd would conduct their arguments with no less passion but with more care. I can't imagine there are many baseball fans who haven't learned just what a gigantic pain in the ass arguing with someone who styles himself a stathead can be.

A favorite contrarian argument I have been encountering lately says that pitchers are ultimately responsible only for whether a batter walks, strikes out, or puts the ball in play. This might be true, but it is an incredibly large claim—one that in any academic setting would require vast scholarship in support of it. I recognize the fun and, often, the necessity of throwing a few (metaphorical) bombs, but if we are to apply statistical techniques to the appreciation of sports, we should likewise apply the attendant restraint and qualification which statistical scholarship demands.

As so many have said, OPS and FIP—newer batting and pitching statistics, respectively—seem to simply be more practical and useful metrics for evaluating success than RBI and wins (although many older stats remain quite useful and accurate descriptors of success). But our insistence on the superiority of some new statistics over traditional alternatives must change to reflect a simple fact: the revolution has already occurred. The new statistics aren't going anywhere. Perhaps the best aspect of sports is the exceedingly rare purity of meritocracy that exists on the playing field; players who can play will do so, those who can't will soon find themselves out of work. So let it be with statistics: those stats that provide practical benefit will endure.

What I am asking, ultimately, is that those of us who are dedicated to further implementing the new statistical models be more gracious than those who originally opposed their implementation in the game. There is a discouraging tendency among most intellectual insurgencies to reject their opposition's beliefs, but to mimic their conduct, and too often, those advocating the new metrics have allowed their opposition to define the tenor of the conversation. We should never be so dedicated to a particular way of knowing that we close our minds to the practical insights of others. We should take care that we don't reach the self-defeating position of pretending that men who played and coached the game for decades—who know how to drop a perfect squeeze bunt or exactly when to pull a double switch—have nothing to contribute that can't be better understood by anyone watching from the stands. And we should take care that we never get to a point where a fan can't say something as simple as "he's really been raking lately" without being challenged to confirm it with a slew of statistics.

The clinical words of analysis, after all, exist to complement and strengthen the words of the visceral and the immediate. Private vocabularies can inform and explain, to those who speak them, and we should be generous in teaching them. But there is no benefit to learning a new language and ignoring the old, and the entire point of including others within our new language is to remove the divisions that might cause us to look down on those who don't speak the way we prefer.

Finally, we must remember: these evolutions are always ongoing. There are ideas about baseball that we now take as self-evident that will one day seem quaint or outmoded to us. Remembering that is the first step towards recognizing that we are not ever possessed of perfect knowledge, and ensuring that we make our arguments with fairness and charity.

The Pastime We Deserve

The Butterfly Effect