What am I thankful for? Plenty of things! But somewhere on the list is pitch-framing metrics, which are just about my favorite baseball thing in this halfway-through decade.
Speaking of which, this came up during my most recent chat:
Q: Has anyone figured out which former catchers (pre-electronic-strike-zone equipment) might have been great pitch-framers?
A: I haven’t seen anyone try. It’s funny, you think, “Well, that’s just not going to happen without the PITCHf/x data. But then you (should) think, 'Wait a minute, maybe it’s just that I’m not smart enough to figure out how that could happen.' ” So I’ll bet that it could happen, actually.
That wasn’t a terrible answer. A better answer would have been, “I’m sure that someone’s actually done some work on this, and either I missed it or have forgotten.” Because that's usually the right answer. And sure enough:
Well of course he did.
@robneyer Re: the Q about catcher framing pre-PITCHf/x, @MarchiMax looked into it before the Indians hired him away: http://t.co/OSK7ZQBVB7
— Ben Lindbergh (@BenLindbergh) November 26, 2014
PITCHf/x data begins in 2008. Max invented a non-PITCHf/x method, applied to Retrosheet’s pitch-by-pitch data that goes all the way back to 1988. To check his work, he used the same method on PITCHf/x-era catchers ... and his results correlated strongly with the PITCHf/x-based results. Which suggests strongly that his method “works”, unless you just don’t believe in pitch-framing metrics at all.
One important difference between the two methods is the distribution of ratings. The PITCHf/x-based numbers are more dispersed: when one considers catcher-seasons with at least 1500 pitches caught, the standard deviation is close to 13 runs for the PITCHf/x numbers and about 7.5 for the Retrosheet ones. That means the Retrosheet-based values (I’ll call them “RetroFraming”) will yield more conservative results.
Given the good agreement of RetroFraming with the PITCHf/x-based numbers, we can move on to showing some numbers going back to 1988, keeping in mind that we’ll less likely see extreme values with this metric.
The takeaway from the tables I’m not going to show you here? Jose Molina’s a pitch-framing god. Which you already knew. But the other big one is something we might have guessed, but never really knew: Brad Ausmus was a tremendous pitch-framer, certainly among the very best of the last 25 years. I was doing this sort of work during nearly all of Ausmus’s career, and I’m sure that more than once I wondered aloud why teams kept giving Ausmus 400-600 plate appearances every season. Of course they told us it was because of his tremendous work behind the plate. Which I bought, but only to a point.
Well, now it makes a lot more sense. And if you really buy into the methodology, you might make the case that Ausmus’s true “value” was maybe twice what the current Wins Above Replacement methods, exclusive of pitch-framing, suggest.
Which doesn’t get him into the Hall of Fame. It’s just interesting, if you’re interested by such things.
What’s next? Historically speaking, one can’t help wondering if there’s something to be done with the pre-1988 numbers, which largely don’t include pitch-by-pitch data. Maybe there’s nothing. But now that we do have some confidence in the existence of pitch-framing as a skill, and half-a-handle on the value therein, we might revisit some pre-1988 catchers with tremendous defensive reputations.
In Craig Wright and Tom House’s seminal book, The Diamond Appraised, Wright introduced the “Catcher ERA” statistic (at least to an American audience, as the numbers had long been tracked in Japan). Semi-famously, the poster boy for CERA was San Diego’s Doug “Eyechart” Gwosdz, who was allowed to start exactly 24 games in his major-league career, perhaps because of his .144 career batting average. But Wright tracked Gwosdz’s entire professional career, and found that whenever Gwosdz caught, the pitching somehow magically improved.
More generally, Wright wrote, “The mark of the master is the illusion whereby balls become called strikes. Learning to catch the ball so it looks like a strike may do more toward preventing runs than throwing out the extra base runner once a week that is the difference between the best- and worst-throwing catchers.”
Craig didn’t have the ability to measure pitch-framing, specifically. Which is why he turned to Catcher ERA as a sort of proxy, with the obvious benefit (or not) of also including just about anything else you think is important, from calling a good game to (yes) throwing well.
Now, of course, we can do a lot better. We can track not just throwing, but also blocking pitches in the dirt and (yes) pitch-framing. Craig doesn’t often write these days about the current game – instead he writes beautifully about the past – but I would love to know what he thinks about the new metrics, and one of these days I’ll try to sign him up for a guest essay.
