from Amazon
Thursday, May 07, 2009
Evidence that Bud Black is a bad manager…
I have been calling for Bud Black’s firing for a while now (as if anyone cares, right?), if only because his team has stunningly underperformed their presumed/supposed talent so greatly over the last 2 years.
I am fairly confident if you looked closely at his decisions that you would find that he is not a good manager, although you would never be able to find how and why his team has underperformed by double digit wins over the last 2 years or so (especially since most of that is probably just plain old bad luck - although you can never know of course).
Anyway, we had a thread about the Padres lineup and how batting Eckstein, who at this point in his career is one of the worst offensive players in baseball, 1 or 2, as he has done all year, probably costs about a win a year, which is not “wood.”
Today, I was listening to the game on the radio, and with runners on first and second, no outs, bottom of the 8th and a tie game, the announcers mentioned that the defense was really playing aggressively for a sac bunt and that it would be hard to execute it. Now, while the decision to sac bunt or not in a sac bunt situation involves game theory, it is “modified game theory.” By that I mean that the offense gets to see how the defense is playing their hand before they make their decision. That is a critical part of their decision-making process of course. The defense is SUPPOSED to play somewhat agnostically so that it does not matter what the offense does - bunt or not. However, if the defense does not, and plays either too aggressively for the bunt or for hitting away, the offense is now supposed to bunt or not bunt 100% of the time or at least some different percentages than if the defense were paying agnostically and perfectly according to game theory.
Of course, the defense may NOT play agnostically if they know or suspect that the opposing manager is “dumb” and will not play optimally, given the defense. For example, if the defense knows that manager is going to bunt a lot (too much) even when they are playing aggressively for the bunt, then it is correct for them to play aggressively. Likewise, if they know that the offense is NOT going to bunt (or bunt less often than they should) even if they are playing way back, then it is correct for them to play way back.
Anyway, although I was not watching the game today, only listening to it on radio, if the defense was in fact playing aggressively for the bunt (first and third charging hard, the “wheel play” on, etc.), OF COURSE it is a mistake for the offense to bunt! You don’t have to be a genius I wouldn’t think to figure that out. One, your success on the bunt is going to go down significantly, and two, when you hit away, you have all kinds of new holes on the infield. Again, does not take a genius to realize that. However, if you are a “chicken**it” manager, you might bunt anyway. I say that because sac bunting when it is not appropriate is a “chicken**it” strategy by a manager. If it fails, he can blame it on the batter. If he hits away and the batter makes a non-productive out or hits into a GDP, the manager gets criticized, post-hoc, for not bunting.
Anyway, you can guess the rest of the story. Black calls for the bunt anyway, by a batter who is not fast (which is always a reason to tend to not bunt as I explain in painstaking detail in The Book), and the runner predictably gets thrown out at third (if there is more than a small chance of that occurring, of course the bunt was probably not correct in the first place).
Anyway, just a couple pieces of objective data that suggest that Black may not be a very good manager…
Hey MGL, in your first Bud Black thread, I criticized Black for batting Luis Rodriguez in the 2 hole. This was your response:
As far as who bats #2 in the order, I have shown time and time again, that without simming a batting order, or at the very least doing a Markov sim of a particular batting order, you cannot figure out in your head a good or bad order. One, it is too complicated, and two, batting orders just don’t make much difference.
For example, numerous times analysts have complained about a poor OBP but fast guy batting leadoff (which many managers like to do, right?), yet when I run it through a sim, it is just fine or even optimal. Probably one of the mistakes that analysts (and some fans like your self) make when criticizing or “analyzing” lineups in your head is not putting enough weight on or adequately considering running the bases (speed) and other nuances that go into a batting order.
I understand your point that I can’t be sure Rodriguez in the 2 hole is a sub-optimal strategy without running a sim. It seemed clear to me that a .300 wOBA at the top of an order isn’t right, but I get it now that even a decision that seemingly “obvious” needs to be confirmed via evidence.
I guess what I’m getting at, is if it was solely my failure to perform a sim before criticizing Rodriguez’s spot in the lineup the reason my criticism was dismissed?
Or could there be some other factor making Rodriguez’s .300 wOBA defensible yet Eckstein’s .310 wOBA is not? You suggested speed may be a factor, yet Luis Rodriguez has a total of 4 stolen bases (and was caught 3 times) in his career. I know stolen bases (or lack thereof) aren’t always indicative of speed (or lack thereof).
As I write this comment, I’m starting to realize that you were likely just trying to impart the importance of using evidence, and not defending Luis Rodriguez’s placement. I think the lesson for me is that the answer lies in the process, not the outcome.
Thanks.