I’m pretty fascinated by the strike zone. More specifically, while the strike zone is supposed to be laid out in black-and-white in the rules, I’m fascinated by the fact that the strike zone changes. It’s there, at the very heart of the game, and it’s inconsistent. It always has been, and that’s something people just deal with. I’m fascinated by the documented realities of pitch-framing. I’m fascinated by the zone changes with the count. I’m fascinated by bad calls, in both directions.
In the past, I’ve looked up a bunch of should-be strikes that were called balls. In each case, I was searching for some kind of explanation. I noticed that on a handful of occasions, there was a runner on base and he had taken off for the next one with the pitch on the way. So the catcher would’ve prepared himself to throw, taking him out of ordinary pitch-receiving position. This got me wondering what happens to the strike zone when there’s a runner on the move. Not long ago, Baseball Savant added a “stolen base attempt” check box to its PITCHf/x search. So now you know where this is going.
In short, my hypothesis is that there are fewer strikes called when a runner is going. Part would be due to the fact that the home-plate umpire would be distracted by seeing a runner take off. More would be due to the fact that the catcher would have to prepare to throw, meaning they wouldn’t exactly be receiving balls quietly. They’d be pulling the balls back to their bare hands, in theory, costing out-of-zone strikes and some more fringy in-zone strikes. Maybe the simplest way to put it is that, according to my theory, a runner taking off turns the average pitch-receiver into Ryan Doumit.
Baseball Savant makes this pretty easy to research. The site allows you to filter by pitches inside the PITCHf/x strike zone and outside the PITCHf/x strike zone. It also allows you to filter by pitch result. I decided to examine called pitches both inside and outside the zone for 2013, while also covering the whole PITCHf/x era. I think that’s actually all the explanation I need. So now here are statistical results.
2013 season, in-zone
- Overall: 90% called strikes
- Runner(s) on: 88%
- Runner(s) going: 85%
2013 season, out-of-zone
- Overall: 14% called strikes
- Runner(s) on: 13%
- Runner(s) going: 10%
2008-2013 seasons, in-zone
- Overall: 87% called strikes
- Runner(s) on: 85%
- Runner(s) going: 83%
2008-2013 seasons, out-of-zone
- Overall: 15% called strikes
- Runner(s) on: 14%
- Runner(s) going: 10%
What’s observed? Sure enough, in each case, there’s the lowest called-strike rate when a runner’s on the move. The zone is already smaller with runners on base, but it’s even smaller with runners in motion. My best-guess explanation is the one I’ve already offered: It’s harder to get a strike called in the zone, because as a catcher, you’re simultaneously catching and turning your shoulders. For the same reason, it’s harder to sell a strike outside the zone. If umpires can be influenced by little motions when a guy’s receiving a pitch, it stands to reason they can be influenced by other motions when a guy’s getting ready to throw to a base. There’s no such thing as receiving with a still, quiet body when someone’s trying to move up 90 feet.
Just to give you some visual sense of what I’m talking about:
I don’t think “ball” is necessarily the default call, and as we’ve seen, most pitches in the zone are called strikes even with a runner in motion. Still, I suspect if there’s any doubt, a ball call becomes more likely. In an ideal world, umpires would consider only the location of a pitch when it’s around the plate, but framing research shows we don’t live in an ideal world. There are other considerations, which is why the strike zone can dip and shimmy depending on the circumstances.
In case you’re curious, the differences look a little bigger if you eliminate pitches in the middle of the zone, since those are basically always gimmes. The differences are never dramatic, but it’s enough to see they’re there. It’s enough to see a suggestion that something is going on.
I think there are two interpretations, and they’re not contradictory. The first is that umpires call a slightly different strike zone when a runner’s on the move, probably — but not definitely — because of the way the catcher has to react. This suggests the movement of the catcher influences the umpire’s decision. The second is that, while the zones are different, the differences are small, and it’s not like runner activity indirectly makes a mockery of the zone. Umpires are mostly consistent in this regard, with room for just a little external influence. It seems true they call fewer strikes when the catcher’s getting ready to throw, but not by enough to complain about. By only enough to think, “Sure, that could be a thing.”
Unfortunately it’s just dawning on me that the data’s imperfect. If a runner takes off for second, but there’s a called ball four, that won’t show up in the spreadsheet. Apparently it also won’t show up if a runner takes off and there’s a called strike three. PITCHf/x tracks steals and steal attempts, but it doesn’t always designate a runner in motion, so I’m now more frustrated than I was 20 minutes ago. I also don’t know what the next step might be, so I’m content to leave it here, if nothing else as an introduction to an idea. I’ve long wanted to know what happens to the called strike zone when a runner is going. I don’t have all the data I want, but at least I’m not completely empty handed. Intuitively, it seems there are going to be fewer called strikes. I don’t yet have a reason not to believe that.
Print This Post