“Sean,” said catcher Isaac Wenrich to pitcher Sean Conroy, the first openly gay active player in professional baseball history, “slow down and let me put a dip in my mouth. That wasn’t a gay reference. I said dip.”
“In a normal workplace, these comments would be grounds for a lawsuit, or at the very least a sensitivity seminar,” write Ben Lindbergh and Sam Miller, the two self-declared “apostles of sabermetrics” who, as directors of baseball operations for the independent Pacific Association Sonoma Stompers, brought Conroy to the team in 2015. But in the locker room atmosphere of a ball club, “they represent a strange sort of progress.” For any player, “it might hurt more not to be mocked, which would mean being beneath the team’s notice.” Conroy’s own reaction: “ ’Keep doing that.’ It’s what makes me feel comfortable.”
Signing the first gay player—a scorecard from his winning start was requested and put on display by the Hall of Fame last year—was the least of Lindbergh and Miller’s accomplishments in signing Conroy. As a player for the Division III Rensselaer Polytechnic Institute Engineers, Conroy had been passed over by every major league organization as a junk-ball pitcher with an odd delivery. What the authors saw in him, however, were stats showing he could get batters out.
Which he did: a solid win for the sabermetric approach that, since the publication of Michael Lewis’s bestselling Moneyball (2003)—and eight years later, the movie starring Brad Pitt as the Oakland A’s data-appreciating general manager Billy Beane and Jonah Hill as a composite of Beane’s geeky numbers guys—has grown famous for urging that game and roster decisions be based on statistical analysis rather than tradition, experience, “gut feel,” and lore.
Part of Lindbergh and Miller’s success as operations executives came from insisting that Conroy, the team’s best relief pitcher, not be pigeonholed as a closer, protecting “three-run leads in the ninth inning.” Instead, he should be brought in to pitch in whatever inning the game was on the line. Other data-based achievements accompanied this one, including signing about 10 “spreadsheet guys” (that is, players no one in the Stompers organization had ever seen), deploying shifts in creative and generally effective ways (such as a first-ever five-man infield), and bringing scouting reports into the dugout, previously forbidden terrain for baseball executives.
The Stompers were 26-11 in the first half of the league’s split season, well ahead of everybody else. “Our goal,” Lindbergh and Miller had told the players, “is to win the first half”—done—”and go undefeated in the second half.” Instead, they went 18-22 and then lost the championship game.
What happened? First, they fired Feh Lentini, the old-school manager who had led them to victory, because he kept stomping away from their advice-giving sessions while shouting things like, “This is just Baseball 101 because you haven’t fucking played it.” Beyond that, most of the position players Lindbergh and Miller found in their spreadsheets turned out to be duds—so much so, they realized at season’s end, that if they had succeeded in substituting their own lineup for Lentini’s, things “would have been significantly worse.” Too late, they came to see that the manager, hidebound and hardheaded as he was, knew some things they didn’t. Urged by Lindbergh and Miller to replace a player with someone they liked better, Lentini explained that a team was more than a collection of individuals: “If changes are made when guys are doing their jobs,” he told them, “then every single person in there starts feeling the pressure because if guys doing the job get released [then] anyone can.”
As Lindbergh and Miller’s failures demonstrate, throwing out the accumulated wisdom of experienced baseball hands is just dumb. But as their successes show, it’s only sensible to add data-based analysis to the traditional baseball toolbox. That’s why they eventually concluded that what baseball needs is not a stale “stats-vs.-tradition debate” but, rather, an ongoing “conversation about the best way to make baseball decisions” that draws on the insights of both. It’s a much wiser judgment than is made by either those who see sabermetric analysis as a hammer and every baseball decision as a nail or those who see statistics as the revenge of the nerds.
Sabermetrics is a path to the truth, but not the whole truth, about baseball. Moneyball, it turns out, wasn’t nearly as on-the-mark as Michael Lewis had us believe. His favorite player among those scorned by the A’s scouts, a squat catcher named Scott Hatteberg with an uncanny ability to draw walks in college ball, ended up seriously underperforming “the scout-certified prospect Carlos Pena,” whom Beane traded so that manager Art Howe would have to play Hatteberg instead. And as Moneyball critic Alan Hirsch has pointed out, none of the eight players Beane wanted to draft in 2002 even made it to the majors.
The undisputed effect of Moneyball on baseball has been to make the game duller. Pitches per plate appearance—the soporific experience of watching a batter watch the pitcher and catcher throw the ball back and forth—have gone up, both to wear out starting pitchers and to increase the chances of drawing a walk. Stolen base attempts have gone down, along with the excitement they generate. Who cares about running when, the data indicate, walks and home runs are what matter most?
As for pitchers throwing complete games—the old normal—that practice has become laughably antiquated, based on statistics showing that, by the third time through the lineup, batters have figured them out. To the extent that Moneyball got some things right, as a fan, I’ll take wrong almost every time.
Michael Nelson, Fulmer professor of political science at Rhodes College, is the author of Resilient America: Electing Nixon in 1968, Channeling Dissent, and Dividing Government.
