Do you remember the end of the book Moneyball?
In 2002 Billy Beane’s A’s won 103 games, the same as the New York Yankees. The Yankees spent $133 million on player payroll. The A’s achieved the same result with $42 million. But the A’s didn’t win in the playoffs, so despite their success in the regular season, it was all a failure. With the season over, Beane slinks back to his office. He knows nothing is going to change and he’s going to have to scale the same impossibly steep mountain all over again. The payroll isn’t going up enough, he’s not getting a raise or a promotion. More of his best players will leave because he can’t afford to pay them. Beane’s reward is nothing. Then he gets a phone call.
The caller offers him a job, a more highly sought after job than his own; one with a raise, a promotion, a much higher payroll, and a better chance to win in the playoffs. Beane takes the job. A few days later he has a change of heart and decides to stay in Oakland with the A’s. That job he eschewed was with the Boston Red Sox. That call was from John Henry.
Beane’s brief time with the Red Sox was over, but Henry’s was just beginning. Although he didn’t get Beane, the path to success in Henry’s mind was clear. Apply the then-nascent field of baseball analytics, the very thought process that helped bring Beane success in Oakland despite their relatively minuscule payroll, to the Boston Red Sox. Build the best analytics team in baseball and the wins will surely follow.
The path to success in Henry’s mind was clear; build the best analytics team in baseball and the wins will surely follow.
Henry didn’t get Beane, so he got the next best thing: his own younger version, 28-year-old Theo Epstein. Unsurprisingly Epstein did many of the things you might have expected Beane would have done. He brought in high on-base percentage guys like Kevin Millar, Bill Mueller, Todd Walker, and David Ortiz. He fired the old school by-the-gut manager (perhaps a year late, but still) and replaced him with someone who was willing and able to incorporate the percentages into his managing. He traded prospects for ace pitcher Curt Schilling, in part by using analytics to convince him a right-handed fly-ball pitcher could thrive in Fenway Park. Oh, and he hired Bill James, the very man whose writing had inspired Beane in the first place.
That was in 2002. Jump forward 13 years and the Red Sox are three-time World Series champions. They’ve also got a meaningfully different front office for the first time since Beane received that phone call. The Red Sox have gone through two GMs in that time, but though Epstein and Ben Cherington were different people (though their voices sounded eerily similar) they both came from the same analytical-bent background, not one that entirely eschewed on-the-ground player evaluation, but one that placed the it next to analytics as equals, two required pieces to complete the championship puzzle.
But then things went awry. The Red Sox finished in last place three out of four years under Cherington, with the final indignity costing Cherington his job. You can certainly point to logical reasons why the team failed on the field and not all of those failures lead straight to bad decisions from the front office, but when Henry looked at the team’s player evaluation processes, he found a fundamental flaw. This led to perhaps the most surprising revelation of the still-young spring when earlier this week John Henry flatly stated that the team had been overly reliant on analytics, and that going forward they would be run differently.
We kind of all saw this coming with the hiring of Dave Dombrowski last season, who is known as a more traditional general manager, but to hear Henry declare it in such stark terms was striking. There is danger in becoming too wedded to any one way of player evaluation and perhaps the Red Sox front offices under Epstein and Cherington had gone too far in one direction. That’s impossible to say from the outside, even when looking at the decisions the team has made. There are just too many moving parts, from player evaluations signings, trades, drafts, and team performance, to look at any organization over a short period of time and make declarative statements about what they do well or what they do poorly. And if there is a scary part of Henry’s statement, that’s it. If the team has moved too far towards analytics, that’s fine, but the answer isn’t to overcompensate in the other direction.
That’s not to say the team won’t win with an analytic-lite philosophy, if that’s their new philosophy at all. Dombrowski has had success in the past with the Marlins and Tigers with exactly this model, building strong teams through scouting that won many games. If the goal is to build a long-term winning organization, it seems pretty clear based on the last two decades worth of data that the smart teams employ both perspectives in roughly equal measure. That isn’t to say you can’t win with just one, or even have sustained success that way, it’s that watching baseball creates perceptions. Analytics challenges those perceptions. It forces people to confront what their eyes are telling them and to explain it with logic and data. Without analytics teams can get caught in a whirlpool of their own unchallenged thoughts and expectations, and that’s a recipe for bad decisions.
Watching baseball creates perceptions. Analytics challenges those perceptions. It forces people to confront what their eyes are telling them and to explain it with logic and data.
Of course, that could happen by using analytics exclusively as well. The challenge the Red Sox face is to learn the proper amount of both and to apply them at the proper time. Perhaps Henry’s statements to the press are just statements to the press and don’t necessarily reflect the reality on the ground. Perhaps Henry thinks what he’s saying is what the fans and media want to hear. But it’s surprising that after almost a decade and a half of analytically driven baseball decision-making, a time in which the Red Sox have won three World Series championships, that Henry has decided to abruptly change course.
The Red Sox do still employ Bill James and they recently created a position for Brian Bannister to analyze pitching using, among other things, analytics, so they certainly aren’t abandoning analytics entirely. Perhaps they’re just being more selective about things. That could be good. But even if the team has indeed grown too reliant on numbers in their decision-making, the mere fact that the Red Sox are moving away from a philosophy that has led to such tremendous success on the field, even if not recently, is concerning. How concerning? As always, the answer isn’t in the results, but in the process.