Tracing sports’ data revolution, Liam Day wonders what it means for masculinity and where we go from here.
Scott Behson wrote a terrific piece a couple of weeks ago on The Good Men Project about why no team had yet signed free-agent pitcher Kyle Lohse, despite the fact he won 16 games last year and sported an E.R.A. below 3.00. (If you haven’t read it and you’re a baseball fan, you really should.)
Behson’s thesis, that Lohse going unsigned represents the apotheosis of sabermetrics, in whose new statistics Lohse doesn’t measure up nearly as well as he does on the traditional measures—wins, E.R.A.—has got me thinking. What does the evolution of data auger for masculinity in the 21st century? Is there something particular in the masculine psyche that’s allergic to data? As statistics come to dominate more and more fields of endeavor, how much control over what we do can any of us say we have? What is the line between science and art in any professional field? What will be next for the Bill James and Nate Silvers of the world?
The data revolution didn’t start in baseball. And it isn’t all that recent a phenomenon. It started after World War II and, to the detriment of American manufacturing, in Japan. For what we’ve witnessed in baseball over the last 15 years or so is merely the logical evolution of the mathematical principles that men like W. Edwards Deming applied to quality control on the assembly line.
As would become the trend whenever data spread to a new field, Deming was practically ignored by American businessmen and managers for most of his life. If he was known at all, he was known as a crank, which is why he needed to go almost half-way around the world to find an audience receptive to his ideas.
Japan was a poor, defeated country after World War II. Deming first visited in 1947, at the request of the United States’ Army, to assist with a census. After an invitation to speak from the Japanese Union of Scientists and Engineers, he spent months training them in statistical process control. So taken with him and his ideas, in 1950 the Union named a quality prize after him.
The result, as if anyone doesn’t already know, is that, between 1952 and 1970, the prices for similar American and Japanese cars trended in opposite directions. In 1952, the price of an American car had been $1,500. By 1970, it was $2,215. In 1952, the price of a Japanese car had been $2,950. By 1970, it was $1,210. A relentless focus on quality helped drive down costs for the Japanese manufacturers. American manufacturers focused on cost at the expense of quality.
To Deming, American managers knew only how to maximize profit. That’s it. As he is quoted in David Halberstam’s The Reckoning, “They know all the visible numbers, but the visible numbers tell them so little. They know nothing of the invisible numbers. Who can put a price on a staisfied customer, and who can figure out the cost of a dissatisfied customer?”
In baseball terms, the visible numbers are those that every boy and girl who grew up following the game know. The visible numbers are the home runs and the runs batted in, the batting and earned run average. The invisible numbers are the ones the sabermetricians came up with: batting average on balls in play (BABIP) and fielding independent pitching (FIP).
From manufacturing the data revolution spread to the world of financial trading, where, in the last decade of the 20th century, during the heady days of the tech bubble, quants wrested control of Wall Street’s trading floors from the meaty hands of those Tom Wolfe, in his novel The Bonfire of the Vanities, so appropriately called the Masters of the Universe.
Wolfe’s enduring theme, across decades of fact and fiction, has always been masculinity. He spies Darwinism everywhere, base biological struggle in everything we do, every interaction we engage in. Returning to the scene of Sherman McCoy’s crime 25 years later, Wolfe wrote a longform piece for Newsweek back in January. He called it The Eunuchs of the Universe.
Anyone looking for a detailed explanation of how the quants became Wall Street’s kings should look elsewhere. However, anyone interested in the anthropology of it, what it means that the nerds and their computers have supplanted final club alumni as the true sources of power on the trading floor, would do well to read Wolfe’s imaginative reenactment.
Heretofore, trading took testosterone. One needed balls to put big money in play. Of course, when Glass-Steagall came crashing down, one didn’t need quite so much testorone or the size stones one has to readjust all the time because they hang so low. Tearing down the wall between the retail and investment sides of the big banks meant traders were playing with house money. But by that time, it didn’t matter. The quants had cut the balls off the traders, relegating them to irrelevancy.
The reversal didn’t happen overnight or without a fight. The same names hurled at baseball’s sabermetricians were hurled at the quants. Quants weren’t real men. They didn’t play the game, choosing instead to hide behind their computers, programming algorithms that would execute millions of trades daily on penny and half-penny margins, recognizing the accrued value of those margins over millions of trades. The big one didn’t matter anymore.
So it was with baseball when the sabermetricians began their assault on America’s pasttime. As Scott Behson pointed out in his piece, sportswriters, players, and fans alike called the new stats-obsessed commentators names: geeks, nerds, basement-dwellers. Like the traders who had already fallen victim to the statistics revolution before them, players and announcers like Joe Morgan were mad, primarily because the sabermetricians had never played the game. They’d never stood in the batter’s box, a mere 60 feet 6 inches from a pitcher throwing 95 miles per hour. What could they possibly know about what should or should not be valued by baseball teams, what could or could not be controlled in the confrontation between pitcher and hitter?
Nevertheless, Billy Beane built a contender using sabermetrics and Theo Epstein a World Series champion. A book was written and then they made the book into a movie. They even got the guy from The West Wing to write the script. It doesn’t get more mainstream than that. The final result is, as Behson pointed out, that no one wants a 16-game winner who has an E.R.A. below 3.00.
From baseball, the stats revolution moved to football and basketball. One started to hear about Yards After Catch and Yards After Contact to measure the effectiveness of receivers and running backs. The Houston Rockets developed a statistical framework that went beyond points, rebounds and assists to measure a basketball player’s contribution to his team.
As the New York Times pointed out back in 2009, the star of basketball’s new statistical framework was then Houston Rocket Shane Battier, a geek in a basketball player’s body. And, unsurprisingly, the response was as it had been in manufacturing and finance and baseball before it. Opposing players “think of me as some chump,” Battier was quoted as saying. “No one dreads being guarded by me.”
But guard people has been precisely what Shane Battier does. Yes, he is 6’8” and, yes, it sometimes seems he is willing to forgo any role on the offensive end of the floor to focus exclusively on defense, but what made Battier an especially effective defender when he was with the Rockets was his statistical knowledge of his opponents, gleaned from the reams of data the team’s front office shared with him.
When guarding a player like Kobe Bryant, for example, Battier knew exactly where on the floor and in what situations Bryant was least efficient and his entire defensive game plan was designed to force Bryant to just those spots in those situations. To put it more precisely, Battier’s game plan was to force Bryant to shoot off the dribble moving left from 18 feet out. That didn’t mean Bryant wouldn’t occasionally make that shot. It simply meant that over the course of a 48-minute game Bryant was statistically likely to make fewer of those shots than any other shots he might take.
This isn’t exactly beat-your-chest, dunk-on-your-opponent stuff. It means that when your teammates are getting amped, listening to Guns’-N-Roses (yes, I know, I’m dating myself) or Eminem (Lose Yourself is one of the all-time great songs to get amped to), you’re sitting quietly in front of your locker studying spreadsheets. In a locker room’s distinctively masculine culture, it marks you as different.
From sports, the statistical revolution moved to politics, because where is more data collected than in politics. Absolutely everything—from what candidates voters choose to what issues candidates emphasize to what sides of the issues they take to what slogans they put on their campaign literature—is polled and poll-tested. As with all things human, polls have biases. Some lean left, some lean right, some are more accurate than others. What Nate Silver did during the last two Presidential elections is simply take the infitude of data the polls were spitting out on an almost daily basis and aggregate them, assigning different weights to specific polls based on their perceived biases and track records for accurately predicting the outcomes of past elections.
What was the result? 50 for 50. This past fall Nate Silver accurately predicted the Presidential electoral result of every state, though predicted is not really the right word for it. Silver was hardly taking a shot in the dark, reading chicken entrails for clues to the future. The data were overwhelming. Anyone who was paying attention at all knew well before November that Barack Obama would be reelected President.
But, again, the reaction of longtime political observers was to attack the new nerd on the block. Silver, who, before moving to politics, began his statistical career as a baseball sabermetrician, was called all the same names his fellow statistical junkies have been called. He was a sissy, effiminate “with a soft-sounding voice.” Joe Scarborough famously called him a joke and then, after the election proved Silver’s methodology correct, only half-heartedly apologized.
By this point, the pattern should be abundantly clear. Data disturb the male psyche. It’s not just fear of the new—new ways of doing business, new ways of playing the games we learned from our fathers—but fear that relying on data, which is, by definition, collective, saps us of our rugged individualism. I mean John Wayne never needed data to get the stagecoach to Lordsburg or win Maureen O’Hara’s heart (multiple times) or track down and rescue Natalie Wood from the Comanche. John Wayne relied on grit and his finely honed instincts.
But that’s the thing about instincts. Instincts are but the translation of thousands of pieces of data into split second reactions, reactions that happen so quickly we’re not even conscious of the data we process to react the way we do.
To extend our Western theme, I’m reminded of the movie Unforgiven, at the end of whose concluding gunfight the less than intrepid reporter, played by Saul Rubinek, asks Clint Eastwood’s gunman whether, of all the men in the saloon, he went for the sheriff first because he knew the sheriff was the best shot. I imagine that if Shane Battier had been a 19th-century gunman facing more than one opponent at a time, that is precisely how he would have approached it.
So what’s the next country to be conquered by statistics? Or is there a limit to the application of data for improving performance? And is six sigma even desirable in all walks of our lives? After all, to err is human. I hope to explore those questions in part 2 of this essay.
Photo: AP/Nam Y. Huh