- FanGraphs Baseball - http://www.fangraphs.com/blogs -

ALCS Coverage: Girardi Is Nuts

There are a lot of reasons the Yankees lost last night, and most of them have to do with bad pitching. A.J. Burnett threw a lot of bad pitches. Phil Hughes threw some bad pitches. It happens. The Angels are a good team, and they pounced on some mistakes. New York is still pretty likely to win one of the last two games and go on to the World Series. In the end, last night’s game is probably a footnote.

However, that doesn’t let Joe Girardi off the hook. I’m not the only one who has called him out for over-managing in this series, but last night, he took it to an incomprehensible level. I’m going to gloss over the fact that Mike Scioscia intentionally walked Alex Rodriguez and put the tying run on base – that’s another post entirely. And while the decision was bad, you can at least kind of understand his thinking there. But then Girardi took Scioscia’s questionable managerial move and raised him all-in by sending Freddy Guzman in to run for A-Rod.

The Yankees were down by a run with two outs, so yes, maximizing the odds of scoring that baserunner is a good idea. However, any move you make in an attempt to improve your odds of scoring has to be proportional to the penalty of making the move. If you’re taking Rodriguez out of a game that could easily go to extra innings, you’re inflicting pretty significant harm to your team later on, so there better be at least some kind of real, tangible improvement in the likelihood of scoring.

Guzman is fast, no doubt, but so is Alex Rodriguez. A-Rod has 297 career stolen bases while succeeding at an 82 percent clip and he was 14 for 16 this year. Yes, he hits home runs, but he’s both fast and good at running the bases. The marginal improvement in speed from having Guzman stand on first base instead of Rodriguez is minuscule. There was essentially one play where the difference in speed could have come into play – a double into the gap or down the line that leads to a play at the plate.

In his career with the Yankees, Hideki Matsui has 207 doubles and triples in 3,820 plate appearances. 5.4 percent of his trips to the plate have resulted in an extra base hit that didn’t clear the fence. Over the last three years, he’s at 4.6 percent. So, let’s just say that the odds of Matsui hitting a ball in the gap or down the line was about five percent, or one in 20.

Now, we have to eliminate all the plays where speed won’t really matter – either runner could easily score on a ball that gets to the wall, given that they’d be running on contact with two outs. Neither runner likely scores on a blooper down the line that just finds the chalk or a ground ball past the first baseman that hooks into the corner. The ball has to get past the outfielders for there to be a play at the plate, but not so far past them that they couldn’t get it back in fast enough to make it close at home.

So now we’re talking about a fraction of those one in twenty odds. The real odds of that one specific play happening? One in 50? One in 75? It’s somewhere in there. And now, when that play occurs, what is the actual improvement in the odds of scoring with Guzman on base? Even if you think it’s a 50 percent improvement, now you’re staring at something like one-in-100 or one-in-150 that the move works out.

Joe Girardi took Alex Rodriguez out of a game that could have easily gone to extra innings for something like a one percent improvement in his odds of scoring. That’s just insane.

This makes replacing Robertson with Aceves look downright tame. Someone stop Joe Girardi before he manages the best team in baseball right out of the playoffs.