THE BOOK--Playing The Percentages In Baseball

I spoke with the author for about 20 minutes on the phone the other day and then e-mailed him a somewhat detailed analysis of that scenario last night.  Maybe my e-mail arrived before he wrote this copy.  The article started out very well and then quickly fizzled.  I am still waiting for Part II which tells us the answer!

I basically explained to him in the e-mail that down by 2 runs, as opposed to tied or down by a run, a bunt attempt is probably nevre going to increase the WE and that this was probably a no-brainer, in terms of bunting or not.

I also explained that this was a unique situation in terms of the decision.  Usually you have a certain batter at the plate and you decide whether to bunt or not by comparing the WE with that batter hitting away and with that batter bunting.  As I explain in The Book, if it is a weak batter (and decent bunter), the infield generally plays in anticipation of the bunt so that the bunt attempt yields a relatively low WE, but happens to be around the same WE as the poor batter swinging away.  With a good batter, the WE from hitting away is obviously much higher, but since the defense doesn’t expect a bunt, they are playing back such that a bunt attempt by this batter (with the infield playing back) actually yields a much higher WE than one by the poor hitter, assuming they are about the same speed and quality of bunters (of course the poor hitters are usually better bunters).  So no matter who is up to bat, if the defense is playing properly (which they often are, more or less), it usually doesn’t matter if the batter bunts or not regardless of the hitting quality of the batter.  Of course, with the defense playing optimally (such that the bunt or non-bunt yield the same WE), the offense must still bunt or not-bunt a certain optimal percentage of the time, otherwise the defense will change its position at a “cost” to the offense.  Of course, for one and only one play, it does not matter what the offense does (bunt or not bunt) if the defense is playing optimally.

Anyway, what was my point?  Oh, that this was a unique situation in that the offense has a choice of batter.  If they are going to bunt, they would have pinch hit Woodward, Hernandez, or even a pitcher, and if they were going to hit away, they would bring in Floyd, their best hitter off the bench, which they did.  Well, that is a whole different ballgame!  Normally, the offense gets to see where the defense is playing and then act accordingly.  If they are playing optimally, then, as I explained, it dosn’t matter what they do, bunt or not, and they must mix it up.  If the defense is NOT playing optimally (either too far back or too far in), then they would either bunt or not, 100% of the time.  In this case, however, the defense is going to be able to tell by who pinch hits whether the offense is going to bunt or not.  Floyd they don’t bunt; someone else they do.  That enables them to either play way back or way up.  (Of course, with the defense playing way back, it may be correct for Floyd to then bunt, but with his bad ankle, that may not be true.) Anyway, because the offense is forced to tip their hand, which is actually not a bad thing, since they indeed have a choice as to not only what to do, but also by which batter, there is ONLY one correct choice, and that is to pinch-hit Floyd and hit away.  In fact, you would NEVER purposely not bat your better batter and put in your worse batter with the express intent to bunt 100% of the time.  The ONLY time you might do this, and it is going to be a rare event, is when even with the defense playing all the way up, the poor batter is a good bunter with speed such that his WE when bunting (and the defense breathng down his neck) is higher than the with the good hitter hitting away.  That might happen if the good hitter is not THAT good, the bad hitter is an excellent bunter with great speed, AND it is like the bottom of the 9th in a tied game.\

I pretty much explained all this to the author in my e-mail.  The summation of the above is that with the option to pinch-hit with a very good lefty hitter like Floyd versus the righty pitcher (and not a great closer), even with a bad leg, and being 2 runs down as opposed to 1 run down or tied, it is not even close.  The bunt attempt by another pinch hitter, with the defense clearly anticipating the bunt, would yield a MUCH lower WE.  Not even close.  I wish the author would have at least stated that I had said that.