Much discussion last night and today about the decision by Dave Anderson to send Mike Napoli home last night, trying to score from first base on a single by Elvis Andrus with two outs in the bottom of the ninth and Andres Blanco coming up next.
Tom Tango has a run expectancy chart that shows that, from 1993-2010, with runners on first and third and two outs, at least one run scored 28.8% of the time. BP has a story from 2009 that shows that the run frequency at 26.7%, presumably because they are using a most recent and narrower span of years (remember that the late 90s and early 00s were eras when offenses performed at historically high levels, which skews the frequency upwards).
You can tweak the numbers if you like...you can push the percentages down because Joakim Soria is a great pitcher, or push them up because Soria was struggling, or push them down because Andres Blanco is a weaker-than-average hitter.
But in any case, the likelihood of the Rangers getting Napoli home if he's held at third in that situation is in the 25-30% range. Which means that, if you are going to send Napoli, he has to have at least a 25-30% chance of scoring for that to be the right decision.
The gut reaction is that sending Napoli is a mistake, that all Kansas City has to do is execute and he'll be out, and you're better off letting Blanco swing the bat. I think part of that has to do with the almost passive nature of sending Napoli...if you are sending Napoli, you aren't doing so in the hope that he'll do something right or make a play. Once you send him, in that situation, you are doing so hoping that the Royals make a mistake in execution. And I think that, on a visceral level, that sort of lack of control, taking the chances of success out of your hands and instead hoping for the other side to make a mistake, just feels wrong. Rely on your guy -- in this case, Blanco -- to make a play and do something good, rather than relying on the other side to screw up.
But assuming Napoli is going to be thrown out is working off a set of nested assumptions. I harken back to the explanation I was once given about why you want to err on the side of betting when you are heads-up in Texas Hold 'em -- for the other player to win, first they have to call you, second they have to have a better hand, and third they have to hope you don't suck out.
Similarly, in sending Napoli, if the Royals execute, he's out. But first, Mitch Maier has to make a good throw to the cutoff man. Second, the cutoff man has to field the throw cleanly and make a good throw home. Third, the catcher has to catch the ball cleanly. Fourth, the catcher has to get the tag down on Napoli.
All those things are much more likely than not to happen...but they all have to happen together. Just to pull numbers out of the air, if each of the two throws have a 95% chance of being made, and the two catcher actions each have a 90% chance of being done properly, that gives Napoli about a 27% chance of being safe...in other words, right around the breakeven point.
Given all that...I am putting up a poll on this issue. Cast your vote below...