June 12, 2014
The Wisdom of Pinch-Hitting with Pitchers
You could say that the most Sergio Romo save of Sergio Romo’s 2014 season was his 17th, a wobbly six-batter adventure against the Cardinals in which he entered with a two-run lead and held on for a one-run victory in this sequence.
RHB Jhonny Peralta: Strikeout swinging (slider, slider, sinker, slider)
Okay, Romo is really good, so maybe that’s an unfair trademark. If you’re a nicer person, maybe his most representative save of the year was an even more recent one—no. 20 this past Sunday against the Mets.
RHB Eric Campbell: Strikeout swinging (slider, slider, slider, slider)
None of this is at all surprising. With his devastating migratory slider, Romo is a beast on right-handed batters. This year alone, righties are OPSing .445 with one walk and 16 strikeouts against Romo, while lefties are OPSing .740 with four walks and five strikeouts against him. In 2013, the split was .511 vs. .745, with a K/BB ratio more than twice as high when the platoon advantage was in his favor vs. when it wasn’t. In 2012, he had a reverse OPS split, but that was all BABIP driven. His peripherals still showed an extreme, platoon pitcher-like 9.5 K/BB vs. righties and 1.5 K/BB vs. lefties.
And it was by the end of the Cardinals save, as Romo stayed true to his established pattern, that I was thinking to myself with total seriousness something I never had before—not even in a joking way.
Isn’t there a left-handed-hitting pitcher available? It would have to be a better shot, right?
The answer in the case of that Cardinals game was “no.” Well, it was pretty much no. Their only left-handed-hitting pitcher is Jaime Garcia, and he had started and been removed from the game already. The only other options were relievers who might not know which end of the bat they’re supposed to hold.
In the Mets game, the answer was yes. There were left-handed-hitting pitchers available. Sure they were left-handed-hitting pitchers from the staff that set a record for a hitless streak to start the season, but they were interesting options. There was Jon Niese, a career .157/.243/.188 hitter, the strange line a result of the second-most walks of any pitcher in the last four seasons (18 to Ian Kennedy’s 21). And then there was Jacob deGrom, who broke that hitless streak and is 5-for-9 with a walk to start his major league career, with a .256/.273/.372 line as a professional.
Even if pinch-hitting a pitcher for a position player were a sound thing to do (and more on that in a moment), there was no reasonable expectation for Terry Collins to do so. That’s not why they’re on the roster, first of all. Somebody could get hurt. And perhaps least important in any tangible way, but most important in Collins’ mind, if this ever entered that mind at all, he would have had to spend 99 more games managing a player whom he pinch-hit a pitcher for as the tying run. That’s not necessarily a guy you want in your clubhouse and under you in the company hierarchy.
The pitcher pinch-hitting for a position player, a sight that could be seen maybe a dozen or two times a year back in the 1950s, is almost extinct. Pitchers have become worse hitters with respect to the league-average hitter, but even the best hitting pitchers aren’t getting to do it anymore.
The only reason anyone’s done it this decade is that in a 2012 Rays-Orioles game (yes, the last time was actually in an AL game), Ryan Roberts got hurt in the middle of an at-bat in the 11th inning and the Rays had to let Chris Archer take over a two-strike count. Archer actually got the count back to 3-2 before taking strike three.
(It was in the strike zone, by the way.)
But not since 2009 has a pitcher been brought in with any forethought for an at-bat, and each of the last two times—in 2009 and 2008—it was Micah Owings of the Reds, once for Danny Richar and once for Paul Janish. Owings, of course, had periods of his career in which he was actually a hitter, as did Brooks Kieschnick and Rick Ankiel, who together with Owings combined for five of the eight occurrences in the 2000s.
Owings for Janish isn’t really one where you need any extreme situations to make it seem worthwhile. Owings was a much better hitter than Janish, without much qualification. He had an edge of .310 to .284 in OBP and .502 (!!) to .288 in slugging, with an advantage in the minors as well (since Owings’ offensive sample is somewhat small). Against right-handed pitchers, Owings had a .917 to .557 OPS edge over the righty Janish. It could and probably should be argued that this swap didn’t happen nearly enough.
But should teams be doing this more often? First, there’s a reason that the example in the beginning was a closer. If it’s in a setup situation, where a pitching change is possible, then you’ve just burned a hitter in favor of a pitcher who might now have no platoon advantage, and it’s blown up in your face. A team would have to be sure the pitcher they see is the pitcher they’ll get.
There’s also the question of whether pitchers get the same platoon advantage that regular hitters get. Where this is intended to work is against closers like Romo who are so devastating against one sort of hitter, but it helps a lot if the pitcher you’re using to pinch-hit also gets his own boost from the platoon advantage.
The numbers on that almost look hand-made. Over the last four full seasons (I swear this part wasn’t hand-picked) pitchers have almost exactly the same platoon advantage as position players, if considered additively.
Platoon advantage vs. RHP
Platoon advantage vs. LHP (always more significant)
If you consider it as by what factor the pitchers improve rather than by what raw amount, then pitchers have an even bigger platoon advantage than position players. No matter what, though, platoon effects still exist when pitchers are batting and should be considered part of the calculation when weighing pinch-hitting with a pitcher, just as you would with any hitter.
Additionally, you would think that pitchers have to be subject to the pinch-hitter penalty, which while difficult to observe (because bench players generally aren’t starting for a reason and are often facing different classes of pitchers), has been the subject of plenty of research with adjustments for those sort of factors.
Should every manager start rummaging through his rotation for pitchers once he gets to the ninth against a closer with a big split? Not necessarily. Should this sort of thing be happening more than once in five years? Absolutely. There are plenty of times when even with that pinch-hitting penalty, the platoon advantage of the pitcher hitting combined with a closer who’s nasty against the handedness of the upcoming batter would make it so that if it’s a situation where if it’s vital to keep the game alive, this would be a better play.
Like so many things in baseball, personalities are probably a huge reason why this doesn’t happen—both the ego of the position player and the disdain the manager will have for facing heat when in all likelihood the pitcher makes an out. We’re only now getting around to the best hitter batting second and the egos on the field and in the dugout being a little more okay with that. This one would take some more time, but at some point, those pitchers who specialize against hitters in one batter’s box will probably start to see pitchers in the other one.
Thanks to Andrew Koo and the Baseball-Reference.com Play Index for research assistance.