from Amazon
Tuesday, September 27, 2011
Catcher Framing Skill, part WOWY
Mike Fast generously posted his file on Catcher Framing in his tremendous article last week. That file is the data source of everything I’m about to post here.
Also note that when I’m going to present a rate stat, it’ll be “per 75 called pitches”, because there’s around 75 called pitches per 9 innings. Since runs allowed per 9 innings uses the notation RA9, I’ll use CP9 for called pitches per 9 innings. (That it bears some resemblance to CP30 is either an unfortunate blemish, or a cool byproduct.)
Another note is that turning a called ball into a called strike is worth roughly +.12 runs. However, the way that Mike has identified the pitches, it’s more like a “probably would have been called ball” turning into a called strike. As a result, the run value, using this data source, is going to be more like .08 runs per extra call.
On to the data.
Brian McCann (id=435263) was behind the plate for 42,098 called pitches. He ended up with +2.4 more called strikes than the average catcher did per 75 called pitches (CP9 of +2.4). Among the 31 catchers with at least 20,000 called pitches, he’s #1 by far. Getting 2.4 more called strikes per game than average doesn’t seem like much, but when one called strike is worth .08 runs, we’re talking about observing almost 0.20 runs per game of savings.
(This is why you can’t really see it in catcher ERA: the difference is so small relative to the possible noise, that you can’t see it. But 0.20 runs per game is actually huge, especially if you catch 125 games, as that comes out to 25 runs. Observed, and unregressed anyway. And of course, biased based on his pitchers, which I’ll talk about right now.)
Derek Lowe (id=117955) was on the mound for 7,944 called pitches. He ended up with a CP9 of +6.0! Among the 163 pitchers with at least 3000 called pitches, Lowe is by far #1. Derek Lowe gets those called strikes that no one else gets.
Lowe and McCann are both on the Braves.
So, is McCann benefitting from Lowe, or Lowe benefitting from McCann? Or both help each other?
Enter WOWY. Several years back, I introduced a process called With Or Without You (WOWY). It involved, not coincidentally, catchers, and their SB, CS, WP, PB, BK, PK. Pitchers and catchers are a two-man team there. WOWY is a good way to isolate the performance of one player, when you’ve got alot of paired dependencies. While the idea is pretty cool as to how it works, and a bit complex in the coding, the lifeblood of the process is the data. In that case, it was Retrosheet. In this case, it’s PITCHf/x data, as processed by Mike Fast. What I’m going to do is nothing without that data.
Anyway, let’s start with McCann. Of his 42,098 called pitches, only 4212 involved Derek Lowe. Together, as a pair, they got a +5.8 CP9. McCann, WITHOUT Lowe, was a +2.0 CP9. Therefore, this data point is telling us that Lowe was worth +3.8 CP9 above the other McCann-pitchers.
If you need some math (and who doesn’t?), it looks like this:
McCann + Lowe = +5.8
McCann + Others = +2.0
If you subtract one from the other, McCann cancels out, and we get:
Lowe - Others = +3.8
That is, Lowe is worth +3.8 above “Others” (which in this case, is all the other pitchers that McCann caught). Given that there were 67 other pitchers, we’re going to presume that “Others” are fairly representative of the league average pitcher.
Lowe however was also caught by other catchers, notably Russell Martin, when both were on the Dodgers. As it turns out, Russell Martin was #2 in CP9 among the regular 31 catchers. How fortunate that Derek Lowe was caught by these two catchers!
Of Martin’s 42,186 called pitches, Derek Lowe threw 2622. Martin, with Lowe, had a CP9 of +6.7! Martin, without Lowe, had a CP9 of +1.3. Therefore, Derek Lowe had an impact of +5.4 CP9.
Lowe was also caught by a few other catchers (total of 1110 pitches). Going through the same process for each catcher, we end up with Lowe being +4.9 CP9 with each catcher, and each of those catchers being +2.2 CP9 without Lowe. This gives us an impact for Lowe os +2.7 CP9.
To recap:
+3.8 CP9 for Lowe (weight of 4212 pitches; inferred via McCann’s other pitchers)
+5.4 CP9 for Lowe (weight of 2622 pitches; inferred via Martin’s other pitchers)
+2.7 CP9 for Lowe (weight of 1110 pitches; inferred via rest of catcher’s other pitchers)
Obviously, Lowe gets favorable calls. The overall weighted average is +4.2 CP9 for Lowe. Remember, his unadjusted observed CP9 was +6.0. After adjusting out for his catchers’ impact, we see that he was instead estimated to be +4.2. Lowe has been caught by good catchers.
We do this for all pitchers. Derek Lowe is still the #1 pitcher. #2, just a sliver behind, is Livan Hernandez. #3 is Jamie Moyer. These three stand out as easily the best in getting favorable calls.
Anyway, now that I have the CP9 for each pitcher, I can then use those CP9 as adjustments for each catcher. For example, Brian McCann has had the good fortune of pitching with pitchers who get favorable calls. Those pitchers have a CP9 of +0.4. It’s not that huge of an impact, but it’s there. So, McCann’s unadjusted +2.4 gets reduced to a +2.0 adjusted. That is, after adjusting for the quality of his pitchers, McCann’s adjusted CP9 is +2.0.
He is still #1 in baseball. We convert that figure to runs by multiplying by 0.08, to give us an impact of +0.16 runs per game (or +20 runs in 125 games). This is all unregressed. However, given the huge number of balls caught, and Mike’s initial estimate that you would regress very little to begin with, this figure will go down by perhaps 10-15% to get an estimate of true talent.
Last among the 31 regulars is Ryan Doumit, with an observed -3.0 CP9, and runs of -30 in 125 games. Given the smaller sample for Doumit, the true talent will be about a 20-25% drop from -30, down to around -25 runs.
Lowe’s an odd case: he throws in the standard strike zone the least of any pitcher in the majors and has done so for each of the last few years. (See my post: http://www.beyondtheboxscore.com/2011/2/18/1994525/missing-the-strike-zone-effectively-the-odd-case-of-derek-lowe)
His pitching style would seem to lend itself quite well to benefiting from a good framing pitcher, as his pitches are consistently in specific areas out of the zone.