When Will Girardi Learn?
The Athletics and the Yankees endured quite an intense battle Tuesday night. The Yankees, after trailing 6-0 going into the eighth inning, scored five runs and ultimately lost by just one run. The game ended with the bases loaded and a fly ball just four or five feet shy of a walk-off grand slam.
The game should have ended earlier. With the bases loaded, two outs, and a full count, a questionable pitch to Robinson Cano was called a ball. A run scored, bringing the winning run into scoring position.
The following graph from Brooks Baseball shows that the final pitch of that at-bat was, in fact, well within the strike zone.
Thus, the game should have ended 6-4, not 6-5, and Nick Swisher never should have come close to hitting his second home run of the game. Regardless, the Yankees lost, and it was largely due to yet another mathematical calamity produced by manager Joe Girardi.
In the bottom of the ninth with runners on first and second and nobody out, Derek Jeter, already 3-for-3 in the game and one of the hottest hitters in baseball, was instructed to move the runners up via the sacrifice bunt.
Jeter’s recent success notwithstanding, this decision was mathematically disastrous. By trading an out for moving the runners up, the Yankees decreased their run expectancy from 1.573 to 1.467.
Nearly one year ago, I wrote about a similar situation in which Girardi called for a sac bunt on a 3-0 count that also decreased the Yankees’ chances of winning. That bunt was the first and last 3-0 sac bunt in all of Major League Baseball in 2010, and also decreased the Yankees’ run expectancy. Not surprisingly, the Yankees lost that game.
It seems there is a growing data base for insane managerial moves by Girardi. My question is simple: when will he learn? The run expectancy matrix is widely available online (see it here, Girardi) and it isn’t hard to find out that 3-0 sac bunts are widely unpopular (see that here, Girardi). Why isn’t the manager of the most popular sports team on the planet aware of simple statistics?