The Data Does Not Show What You Think It Shows
By Thus we get the Boston Red Sox’s John Henry, champion of data-driven rational decision-making, dumping a two-time championship manager because a team projected to win 94 games won only 90—and then injuring himself falling down on his yacht.
Actually, if the sabermetric evaluation of managers has yielded any clear finding, it’s that the vast majority of managers lose effectiveness over time. Even mediocre managers often get teams to perform above expectations for a year, but virtually all then backslide. To give Francona a second decade you’d have to be betting that he’s one of a rare class of managers who can sustain excellence over the long haul. It’s not an insult to Tito to suggest that he’s not one of them, particularly given the unique challenges and pressures of the Boston fishbowl. I’ll leave this for a separate post, but Dick Williams was a brilliant manager, a deserving Hall of Famer, but you wouldn’t want him managing your team for 15 years. Even Joe McCarthy only lasted more than a decade once, and he had a talent edge in that job that even the contemporary Yankees would envy. Francona is a player’s manager rather than a hardass like Williams, but low pressure carries its own perils, and the presiding over by far the worst September collapse in baseball history (on a team with some conditioning problems and atrocious pitchers willing to show up the manager on the mound) suggests that Francona’s time has run its course. And while I like Tito and think he did an excellent job on balance, let’s keep it in perspective. This wasn’t Williams ’67 — he took over a team than an utter numbnuts had one bloop double away from the World Series.
The Red Sox have made some mistakes (hello, Carl Crawford!) that seem contrary to sabermetric principles, but however it works out firing Francona isn’t one of them.