THE BOOK--Playing The Percentages In Baseball

I think this is a great idea and a well presented article.  I was planning on doing something like this with Burnett, after Girardi kept playing Jose in his starts (even in the playoffs!) to see if there was any difference compared to Posada, but I never got around to it.  Max’s reason as to why he didn’t look at Burnett is valid also.

Wow. On 3-0 counts fastball was thrown 100% of the time (for Posada and Molina at least).

Wow, great, great stuff!

I perused the article at 5:00 in the AM.  It looks to me like Posada is more predictable and conventional than Molina so it is not surprising that he would get worse results on his pitches.

Posada likes to throw fastballs (2 and 4 seamers) on hitters counts and more change-ups and sliders on pitchers counts.  Posada’s game calling is more like that of the Little League catcher while Molina’s seems to be more balanced (does not change as much with the count).

It seems to me that Molina is the much better game caller.

I am surprised that there is so much difference when the pitcher is in charge of the pitch selection and Sabathia is no rookie.

Great, great stuff! Did I say that already?

Another great article from Max.  I left a comment at THT asking about any differences in the quality of the opposition batters that the different catchers faced. 

Max is one of the premier Pitch f/x analysts.  It would be interesting if he gets hired by a team to do analysis on American baseball from a continent away.

Very interesting.  My few comments (some of which were recognized, but then not explored, in the article).

1. With the recognition that it was not until the second part of the year that Posada and Molina split time on CC AND that CC was better in the second half, why not do the results comparison only from Start 18 on?

2. By using run value, the comparison is highly contextual (park, batter quality, base-out).  Given the very small sample size, I think these need to be controlled.

I think everything above the “outcomes” section was fascinating, but I don;t think the data lends itself easily to predictably quantifying the outcome.  It was greatly surprising to me that the catcher had such a large control of the pitch selection with someone as experienced as CC. Otoh, this might be idiosyncratic to CC.  Perhaps he has simply always relied on the catcher/coach for that part of the game.  May it would be very different with a different pitcher.  Pure speculation, obviously.

How do folks reconcile this data with the inability of past studies to find any evidence for catchers’ ability to impact ERA?  Have there been any WYWY studies of catcher ERA/defense?

At the moment I am coding WOWY studies of stealing and WP&PB for my Oliver projections.

As I was matching up baserunners with catcher/pitchers pairs, I realized that I could do the same, matching catchers with pairs of batters and pitchers, to see how much each catcher influences the outcomes of the plate appearances.

Coming soon…

Brian: the one thing to be careful about is aging.  I’m working on that as well.  So, make sure to account for the age of the batter or pitcher.