Sabermetricians, like all nerds, must accept that they've won

I'm not going to excerpt too much, because its a paid subscription service, but today on his (excellent) daily baseball missive The Slurve, Michael Brendan Dougherty writes

Numbers aren’t overrated in baseball, but it is easy for a sabermetrics guy to overrate them. To overrate them is to highly rate his own insight and value. "You missed the revolution," [Brian] Kenny spat in a revealing line. Like all totalizing revolutions, sabermetrics in its purest form is dedicated to an abstraction: achieving the most leverage in any given baseball situation. It’s all about increasing the odds of winning. How could anything else be important?
But it isn’t just some fustian and specious baseball religion that stands in the way of a more rational game. Human nature stands in the way. The history of the game and the psychological expectations built by that history stand in the way.

Managers, players, and GMs often have interests that conflict with an ethos that only seeks to increase the odds. And this is natural. While positing a mysterious quality like “The Will To Win” is shamanic nonsense, it is manifestly not nonsense to suggest that some players (like pitchers) thrive on routine, perform best when their roles and expectations are tightly defined, and that they desire to obtain a measurable stat that signals their success - even if that stat’s origins and purpose seem irrational to an Enlightened few.

It happens that I wrote about these issues at length at Wunderkammer Magazine a few years back, which Dougherty was kind enough to link to recently. Were I to write that piece now, I would be far less charitable— to my fellow stat heads.

I continue to find sabermetrics to be a superior way to assess player and team performance in baseball. I haven't wavered in that opinion at all. In fact, I'm in a simulation baseball league (which eats for too much of my time) that is absolutely dependent on advanced metrics. I literally stare at spreadsheets to make decisions in drafting, trades, lineups, defensive shifts, etc. I am not a recent convert. Having come to sports fairly late in life, at least compared to my American peers, I didn't need to be dragged to the new kinds of analysis. I maintain a more skeptical approach than some others. I find that the advanced metrics in basketball and football are not nearly as convincing as those in baseball, which has so many aspects of built-in isolation that it makes separating variables easier. I find the predictive ability of advanced metrics far overblown, and indeed, every year those aided with sabermetrics make wildly inaccurate predictions about team success or failure. Finally, the affectation of some to take about sabermetrics as a science is aggravating if only for its inaccuracy. Science requires experimentation. I doubt MLB is gonna let Bill James take calipers onto the diamond during a game anytime soon.

But, look, in the important areas: OBP over batting average, FIP over saves, WAR over RBI. That stuff seems like commonsense to me, and despite their well-manicured pose of being an underdog insurgency, statheads have won and will go on winning.

Jesus Christ, though, some statheads make it hard. This is one of those tough issues where you have to talk about the broader community, rather than just the prominent writers, which can invent talk of strawmanning. Though it's not always the case, those who write for major publications like ESPN.com tend to express themselves with more reservation and charity than the "enlightened fans," the self-appointed arbiters of how to like sports that seem to have fanned out into sleeper cells across the country. And these people tend to express themselves with maximum certainty and minimal charity. It can make liking sports into a huge pain in the ass. I cannot tell you how many times I have encounter people who use advanced stats to make what is to me the correct argument in a way that is tone deaf, or aggressive, or self-aggrandizing, or in other ways a rhetorical failure. I was at a bar watching the Bulls-Nets series last week, and one of the guys I was with talked about how the Bulls are a mentally tough team. Then another guy who was there launched into a lecture about how talking about mentally toughness is the kind of evidence-free traditionalism we have to evolve past and blah blah blah and it was just like, for fucks sake. Nobody at this Buffalo Wild Wings is impressed with your commitment to evidence-based basketball, friend.

That story is also indicative of the way that some statheads misrepresent actual matters of controversy. As someone who reads and conducts educational research in his day-to-day life, I'm frequently confronted with a very seductive, corrupting empirical failing: the tendency to think "that which I cannot measure is not real." This is the perfect way to describe "chemistry" or similarly intangible aspects of social interaction between athletes and coaches. To many in the advanced stats community, any reference to chemistry or similar is disqualifying. But look: human psychology is real. Human social interaction is real. Both have material expression in the real world. To forbid the consideration of them because we lack the appropriate metrics is exactly the kind tail-wagging-the-dog mindset that often renders empirical researchers myopic. The material world has resisted our efforts to quantify all of it for the entire history of science, and in most venues, we can conduct experimentation. That will never be true of pro sports. Accept that there are subjects on which we will never possess perfect information and about which we must muddle through.

I have written, at length, about the fact that nerds have utterly won the pop culture wars and yet continue to posture as though they are a derided minority. The result is totally unseemly: the most powerful force in media and entertainment whines about oppression at the same time as it bullies everyone else. In fact, throughout our culture, the nerds have been winning, in political forecasting and in banking and in positive portrayals in movies in television, yet the expression of grievance from within has not stopped. The dynamic is becoming similar in the world of advanced stats, which are coming to dominate both the front offices of professional sports leagues and the culture of sports media. The result is a perpetual sneer, the ugly aesthetics of people who are winning looking with contempt on those who are losing. What the advance metrics crowd must decide is, do they want to continue to win, and see their perspective gain greater dominance? Or like the fanboys, were they actually more psychically comforted by the pose of being part of a superior subculture? It's an open question.

What I would like very much to say to the Brian Kenneys and David Schoenfields of the world is this: you have to do more to impress mutual respect and epistemological modesty onto the broader worlds of fans. The message that gets conveyed to the masses and then repeated is always more crude than the one you intend. A more sober, self-skeptical attitude has to be instilled in the world of sabermetrics, for reasons both analytic and social.

Incidentally: reading The Slurve is now one of my favorite parts of my morning routine. It's natural, when you're in capitalism, to pay someone for the work they do that you enjoy. Please check it out and consider subscribing.