THE BOOK--Playing The Percentages In Baseball

I don’t subscribe to BP, so I couldn’t read the interview, so could you explain a bit please. However, that lead-in makes me suspicious right off. Baseball common sense? What is that? Common sense is generally accepted lore of a community. Hence, it is often mistaken. Also, I read a paragraph of the interview on the BTF site and I got the impression that Maddon might be one of those manager’s whose idea of using statistics is to start player X because he’s 6 for 10 in his career against pitcher Y.

Yes, that is correct, about the 6 for 10.  They actually have James Click, a former BP writer, working for them, so presumably they are somewhat into “legitimate” analysis, but in order to implement optimal strategies on a day to day basis, you basically have to have a manager who is willing to be “re-programmed.”

Click on my name for one quote from the article.

The comments by Maddon are not all that bad.  I’ll reproduce the quote Tango refers to and comment on it:

Andrew and James Click supply me with a lot. I get the regular packet on a daily basis, and I go to ESPN.com and look at what’s presented there. Then, Clicker presents me with this analysis based on groundball and flyball percentages, like, is this guy a groundball or flyball pitcher, and do hitters with a bit of an uppercut maybe have a better opportunity to hit against him than someone who is more of a flat-swinger. This is the kind of stuff I’ve paid attention to in the past, but now the information is there to look at, and it’s backed up by numbers. So I might make a decision of who to play based on whether someone appears (on the printout) in blue, or if they appear in red, which is a negative, or in black, which is more neutral. Then I’ll try to read into it deeper to see if there’s anything I can use to exploit a matchup. Another thing I’ll do is look at the opposing pitcher to see how he’s been doing recently, and sometimes I’ll look at box scores to see how he did in right-on-right, or right-on-left, matchups against certain hitters I’m pretty knowledgeable about. I’m telling you man, when I’m trying to set this thing up on a daily basis, I’m looking at a variety of sources of information. I’m always looking for an edge. My mind never really shuts off. ...

The thing about the G/F percentages of the pitchers and batters has merit of course.  As we showed in The Book, there is a significant “platoon” advantage for GB batters facing FB pitchers and vice versa.  The problem is that it is hard to take advantage of it, as most pitchers and batters are not extreme flyballers or groundballers, and there is no real advantage to the batter or pitcher unless both are fairly extreme.

Now, whether Maddon actually uses the “G/F platoon” to make decisions is another story. I am skeptical. The other stuff he talks about is mostly crap.  Box scores?  That is manager-think for you.

If I’m not mistaken, Maddon there admits to just about every cardinal sin of Sabermetrics there is: 1) Small sample sizes, 2) Using ESPN for anything other Bill Simmons columns, 3) Looking for the “hot hand”, 4) And being willing to throw it all out if he “just has a feeling about a guy.”

Add in the quotes about “I’m really into fundamentalism” (I know I’m taking that way out of context) and the fact that he gave 110% in a real team effort to insert every cliche in the book into his speech.  *Sigh* We have so much to do.

The thing is, asking a manager to implement a sabermetric strategy is like asking me to build a house.  If you tell me what to do, I am pretty sure I can do it, but I am not going to be able to give you a comprehensible speech on how to build a house that makes any sense at all.

There are two kinds of people in this world.  One, the ones that know what they don’t know (and presumably know what they know), and the ones that think they know things they don’t.  I am pretty sure that no matter what, for all of the important things in life, I want the former.  Most people (and overwhelming majority I think) fall into the latter.  Almost every manager and commentator in sports REALLY falls into the latter.  Scientists, and I use that term loosely, by definition, fall into the former, which is why almost every important question gets answered by a scientist, or at least by a person that falls into the firts category.  Not that we don’t have any uses for persons who fall into the latter category.  We do.  I’m just not sure for what.

Look at Maddon’s handling of the top of the tenth inning tonight. After a leadoff single, Akinori Iwamura (pretty good LH hitter) comes up against Rob Bell (atrocious RH pitcher). He bunts Iwamura thus reducing win expectancy by 2.7% (source: The Book’s WE table, p.43). Of course, that 2.7% is based on an average situation; in this case, it’s actually worse than that since the D-Rays had a stark advantage in the batter-pitcher matchup.

In the bottom of the tenth, he makes the classic mistake of not using his closer/best reliever on the road in a tie game.