May 19, 2014
The Way in Which the A's Are Still Old-School
For the most part, pitch receiving operates on a level that’s easy to overlook. Over thousands of pitches, certain catchers establish an edge, and those edges add up in a way we can’t see without looking at a leaderboard. Every now and then, though, framing on a small scale comes to the fore, usually when it leads to a larger event. Brett Lawrie, let’s say, strikes out looking out a pitch that appears to be outside, hurls his batting helmet at the home plate umpire, and gets ejected from the game. Our first impulse, like Lawrie’s, is to blame the umpire who blew the call. After reviewing the video, though, we realize that the real culprit was Jose Molina, in the catcher’s box, with the catcher’s glove. The ump was a red herring, a patsy, or maybe an unwitting accomplice.
One such sequence occurred in Saturday’s game between the A’s and the Indians, with Scott Kazmir on the mound in a scoreless tie. We pick up the action with one out in the second, no one on, and a 1-1 count on Asdrubal Cabrera.
Thanks to the work of Harry Pavlidis and Dan Brooks, we can calculate the expected outcome of any called pitch, factoring in the location, the count, the pitch type, and the handedness of the batter and pitcher. There was a 57.4 percent probability that this pitch would be called a strike. Pitchers, of course, make their own assessments of strike probability on the fly, drawing upon a mental database of previous pitches and results, and Kazmir’s body language suggests that the number he had in mind here was at least as high as BP’s. The changeup was more or less where he wanted it, but catcher Derek Norris stabbed at it and very clearly pulled it back toward the plate.
Fast-forward to the 3-1 pitch:
Here we have a 65 percent probability strike, slightly closer to the center of the zone; the extra eight percent is enough for Kazmir to place his hands on his hips (but only briefly). Again, the delivery is where it was supposed to be, but Norris doesn’t give home plate umpire Jerry Layne a great look.
This pitch has a 31.5 percent probability of being a called strike, but this time, the call goes Kazmir’s way, maybe because of the “make-up” call effect that Noah Woodward wrote about today. Next pitch:
And a closer look:
Kazmir is charged with his fifth wild pitch of the season, which ties him for third-most in the majors (Oakland’s Sonny Gray is tied for second with six), but it’s a ball that Norris could have had without making an extraordinary effort. However, he doesn’t get his glove down, and the Indians take a 1-0 lead.
After another ball, Kazmir throws this 2-1 sinker:
It’s another pitch that should have been, and usually would have been, a strike: 64.4 percent of the time, to be overly precise. None of these balls has been a no-doubter down the middle; instead, Kazmir is suffering a death by a thousand stabs, bleeding bits of strike probability on most of the borderline calls. To make matters worse, everyone to the left of the KeyBank sign is laughing at him.
Last one: another sinker, same plate appearance, this time on a full count.
This time Kazmir’s hands stay on his hips, and he adds a derisive nod. That’s what the reaction to a ball call on a pitch with an 82.4 percent strike probability looks like. Actually, it gets worse: Kazmir says something he shouldn’t, and Layne almost immediately ejects him from the game. Oakland’s starter is out of the game, with 23 outs left to record.*
Kazmir deserves some of the blame, for giving Layne a reason to run him. Layne deserves some of the blame, both for making more than one questionable call and (perhaps, depending on what Kazmir said) for being a touch too sensitive about the player’s reaction. Norris isn’t innocent, though. Yes, he was set up inside, and yes, Kazmir missed his spot. But shaky receivers make the misses more obvious. Good catchers are quiet. Norris’ glove is like the Manhattan guy at a party in Brooklyn who complains loudly about how far he had to go to get there.
“This is a pitch down the middle,” former catcher Ray Fosse says on the Oakland broadcast. “And the fact that Norris just moved to his right, he called it a ball.”
The intriguing thing is that for the A’s, this wasn’t an isolated incident. Norris and John Jaso, the A’s lefty-righty, behind-the-plate platoon mates, are both below-average blockers. Jaso’s framing, over the course of his career, has rated -17.6 runs per 7,000 opportunities (roughly a full season). Although he’s caught just over half as many pitches as some starting catchers, he’s at the bottom of the leaderboard for 2014. Norris, whose receiving work has been a positive in the past, has also been a below-average framer thus far this season. Largely thanks to those two, Oakland’s staff has lost the second-most strikes (122.5) and tied for the second-most runs (10.7) in the majors, behind only Minnesota’s.
It’s one thing to find Minnesota (led by former A's catcher Kurt Suzuki) at the bottom of that leaderboard. We’re not used to seeing the Twins at what statheads would consider the cutting edge; this is a team that’s still striking out fewer than six opposing batters per nine, and that until recently paid Ryan Doumit to do something other than DH. The A’s, though? It’s surprising not to see them at the forefront of the framing movement, along with the Rays, the Padres, the Yankees, the Red Sox, and the Brewers—who, while they may not have the same saber-savvy rep as those other clubs, were one of the first teams to embrace the shift. (Also, they're hiring!)
Over a decade ago, Oakland's plan, in greatly simplified summary, was to pursue Jaso-like sluggers who drew walks and had high on-base percentages, even if it meant sacrificing defense. That’s not who the A’s are today. On-base skills aren’t undervalued anymore, so Billy Beane has moved on to more fertile fields: depth and roster redundancy, multi-position players and platoons, expensive relievers (?), and defense. The 2014 A’s lead the majors in Defensive Efficiency, both Park-Adjusted and otherwise. They’ve signed or traded for great fly-catchers like Coco Crisp, Craig Gentry, Josh Reddick, and Sam Fuld, and they’ve given plenty of playing time to weak-hitting glove-first guys like Eric Sogard whom FireJoeMorgan would have made fun of. And yet, for all their emphasis on defense, they’ve seemingly missed the boat on one of the big sabermetric advances of the last several years—a still-undervalued skill that until recently was entirely unquantified.
Here’s the big “but": big ballpark and all, A’s catchers have hit .318/.403/.484 this season, the second-best collective catching line of any team. Although borderline calls that go the wrong way hurt the team and clearly frustrate pitchers—had Kazmir taken truth serum, he might have had some words for Norris as well as for Layne—a bat like Jaso’s makes up for a lot of lost strikes.
Beane, like any good GM, can hold more than one idea in his head, and because Moneyball isn’t a monolithic philosophy—we want guys who get on base, no, we want guys who catch the ball, no, we want guys who—the A’s can go heavy on defense at certain positions and on offense at others, depending on where the deals are.
Framing hasn’t been a widely acknowledged skill for so long that teams have had time to tear down their rosters and build them back up again; Milwaukee, for instance, drafted Jonathan Lucroy in 2007, so while we (and they) now know how much they’ve benefited from his receiving skills, it would be misleading to tie his selection to a PITCHf/x study. The A’s acquired Jaso at what seemed like a fairly low cost (though two of the players Oakland gave up for him, Ian Krol and Blake Treinen, are in the majors, and the third, A.J. Cole, is a good prospect who’s off to a strong start in Double-A), and he’s making $2.3 million to mash. For now, that makes him worth playing, even though he has only old-school saber cred, not the new, hip, catcher-framing kind. If he gets too expensive as he progresses through the arbitration process and nears free agency, or if the A’s see an opportunity to save some runs behind the plate and have him DH more often, they’ll pivot to another player.
It’s not apparent that the appreciation of receiving skills has progressed to the point that offense-first catchers are undervalued—that in a rush to recruit the Jose Molinas and Rene Riveras of the world, teams have forgotten how good it is to have some stick at a traditionally less potent position. Still, sometimes the sexy way to save runs is no more efficient than the old-hat way to add them, and efficiency is all that matters. Even if—no, especially if—you’re the original sexy sabermetric team.
*Dan Otero gets 11 of those outs without allowing a run, and Fernando Abad, Luke Gregerson, and Sean Doolittle combine to finish off a 6-2 win. “The A’s have a pretty good bullpen,” the box score says.