Framing the Hitters

All anyone can talk about these days, some of the time, is the matter of pitch-framing. It’s a concept we’ve been vaguely aware of since childhood, or since whenever we started paying attention to baseball, but for a while it was one of those things we ignored because we didn’t know how to measure it. Then some people started to measure it, and it seemed to make a big difference sometimes, and that’s exciting, and people got excited. People remain excited, since framing is a new field and it’s fascinating to learn how some catchers can do it while other catchers struggle. It’s a very small part of the game overall, but it still has that new-stat scent, and evidence suggests at the extremes it’s pretty significant. I’m thankful for the advances in pitch-framing research.

When people talk about framing, or receiving, though, they talk mainly about the catchers. That’s fine, the catchers are most responsible. They’ll talk a little about the pitchers, and that’s fine too, because catchers need the pitchers’ help. It’s hard to frame a pitch thrown to the opposite side of the plate. But good framing has victims, and worse framing has beneficiaries. The pitchers are affected but the batters are affected too, and it stands to reason framing isn’t completely independent of the guy in the box. Batters probably have some effect, so those batters are worth investigating. Which batters end up with the most extra strikes? Which batters end up with the fewest?

I’ll tell you right now: when I’ve looked at this before, I’ve discovered the obvious. There’s variation among the batters, but there’s more variation among the pitchers and the catchers. Batters have only a little to do with the way pitches are called. But as long as we’re investigating a small part of the game, we might as well investigate a small part of a small part of the game, because where do we stop? Why should we stop? It’s perfectly fine to learn for learning’s sake.

This explanation again. I don’t know how many times I’ve gone over Diff/1000 on FanGraphs, but I might as well do it once more. Diff/1000 is a stat I made up. It uses stats available here at FanGraphs — strikes, pitches, Zone%, O-Swing%, and Swing%. Using the plate-discipline data, we can calculate a number of expected strikes (Zone + O-Swing). We can then subtract that from the number of actual strikes, and put that on a per-1000-called-pitches basis. Then we can adjust it to set the league average to zero. For hitters, a positive Diff/1000 means they’ve had more strikes called against them than you’d expect. A negative Diff/1000 suggests a smaller zone. One of these days, you’ll all be familiar with this and I won’t have to keep going over this same summary.

I calculated Diff/1000 for all batters in 2011 and 2012. I looked only at the batters who faced at least 1,000 pitches in each season. The resulting correlation is 0.48, which suggests that we’re measuring something real. A total of 215 batters were investigated. Here’s a table of quintiles, which should also be informative:

Quintile 2011 Diff 2012 Diff
Group 1 25 9
Group 2 10 7
Group 3 0 1
Group 4 -9 -10
Group 5 -26 -14

The top 20% in Diff/1000 in 2011 saw an average of 25 extra strikes per 1,000 called pitches. The next season, they came in at +9. The bottom 20% came in at -26 and -14, respectively. It’s apparent that there’s regression, but it’s also apparent the regression isn’t 100%. Something is going on, here, that has at least something to do with the guy with the bat.

That’s all background. What’s most interesting right now is data for the current regular season. Which hitters have seen the most extra strikes, and which have seen the fewest? I’ve prepared a couple top-10 tables, and we’ll start with the unfortunate hitters. These hitters have seen a higher rate of strikes than you’d expect based on the rest of their numbers.

Name Pitches Diff/1000
Marwin Gonzalez 527 52
Jean Segura 886 51
Clint Barmes 562 50
Yuniesky Betancourt 688 48
Will Middlebrooks 763 48
Michael Bourn 578 45
Paul Goldschmidt 1071 45
Brendan Ryan 618 43
Ian Desmond 832 40
Marcell Ozuna 522 39

Obviously, it’s far from a death sentence. Goldschmidt’s been amazing. Segura’s been amazing. Bourn’s been good. This has only a little to do with overall productivity. But it isn’t irrelevant. Here’s the other side of the leaderboard:

Name Pitches Diff/1000
Andre Ethier 818 -65
Chris Iannetta 728 -51
Alexei Ramirez 771 -44
Carl Crawford 728 -42
Tyler Flowers 541 -41
Nick Punto 659 -39
Miguel Montero 848 -38
Dayan Viciedo 558 -38
Ike Davis 777 -36
Adrian Gonzalez 799 -36

Ethier’s been pretty bad. Davis has been pretty bad. Montero has been pretty bad. Getting extra strikes doesn’t make you bad, and getting extra balls doesn’t make you good. But still, it’s of interest to explore the extremes, those being Ethier and Gonzalez.

Thanks to Texas Leaguers, we can check out their respective 2013 called strike zones:



That’s Ethier on top and Gonzalez on bottom. You might have to stare for a while to see the differences, but the numbers above say they’re there. According to StatCorner, pitches taken in the zone by Ethier have gone for balls 23% of the time, and pitches taken out of the zone have gone for strikes 6% of the time. Pitches taken in the zone by Gonzalez have gone for balls 11% of the time, and pitches taken out of the zone have gone for strikes 10% of the time. Differences exist, although I’m not entirely certain why. A couple batting screenshots:



I don’t yet understand why some hitters get bigger zones and some hitters get smaller zones. It might entirely have to do with a combination of batter height and unbalanced scheduling against various catchers. One notices that there are four Dodgers in the lower table above. One notices that the season is only a third of the way over, and these numbers could look very different at the end. There will be regression and shuffling and the final tables will say different things. But no matter the effect a hitter actually has, some hitters do end up with too many strikes, and some end up with too few. It matters only some, and not a lot, but that’s never stopped us from discussing something before. How many articles have there been about optimized lineup construction?

See, it’s not that Andre Ethier’s been unlucky. It’s that he hasn’t been lucky enough.

Print This Post

Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.

14 Responses to “Framing the Hitters”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Andruw Jones says:

    What about unbalanced umpire exposure?

    Vote -1 Vote +1

  2. The Humber Games says:

    How often does MLB rotate umpire crews? If so many Dodgers are outliers, couldn’t that also mean that their umpires are just bad at calling balls and strikes?

    Vote -1 Vote +1

  3. Skin Blues says:

    This doesn’t prove that “some hitters get bigger zones and some hitters get smaller zones”. You’d need to look at actual Pitch f/X location data to determine that, similar to how Edge% is calculated.

    We can’t just assume that all non-swung-at balls outside the zone are distributed normally for all hitters. We don’t know how much of the effect we’re seeing is the result of some players swinging more/less often at borderline pitches. Maybe guys that have very low Diff/1000 chase anything close to the strike zone, while guys with a high Diff/1000 chase sliders in the dirt and take a lot of borderline fastballs.

    Vote -1 Vote +1

    • There is only a very weak correlation between Diff/1000 and O-Swing%, and I believe previous research that tried to identify guys who swung at wild balls and guys who swung at close balls came up with basically nothing. This was a few years ago. THT?

      Vote -1 Vote +1

  4. Tyler says:

    Anecdotally, it seems like guys that crowd the plate (especially lefties) don’t get the inside strike called on them nearly as often as guys who stand farther off it, presumably because from the umpire’s perspective if it comes close enough to hitting a guy that he has to bail it can’t be a strike. Any way to test this?

    Vote -1 Vote +1

  5. BX says:

    Could division played in potentially matter, especially with unbalanced schedule? (although trends don’t show up in this bunch)

    Because the AL West is jam packed with not-good defensive catchers, while the AL East has some of the best pitch framers around.

    Vote -1 Vote +1

  6. FutilityInfldr says:

    I am pretty curious about this small game inside the game. The inside strike on a lefty and the Maddox-esque extension of the zone on a righty. A low curve ball called strike on a batter in the rear of the box vs a high fastball in a guy situated differently . . . There might even be correlation for a player with an overly “traditional” stance (perpendicular to the inside chalk, directly at the plate, arms-length distance from the opposite side of the plate, neither crouched nor upright) getting fewer framed vs a non-traditional as the umpire’s frame of reference points (in order: release point > knee / plate / belt > catcher’s glove) require less guesswork . . . fascinating to me.

    I know catchers’ stance location changes based on proximity to the batter’s location in the box (even if only so slightly) to cheat every inch they can . . . but Pitching (and locating pitches) is just so damn hard!

    I can’t get enough of this series. Thanks again, Jeff.

    Vote -1 Vote +1

  7. You're no Jack Kennedy says:

    Seems to me that this ignores the biggest thing: are the pitchers actually throwing strikes?

    If yes, you will have a greater number of called strikes. If not, you won’t.

    So maybe the people in the extreme quintiles are just those who pitchers consistently pitch in the same way. Batters who you pitch around or who have high chase rates don’t induce the pitcher to try to put balls in the zone. Batters who can’t hit or have a good eye would induce more.

    The Zone + O-swing blending that happens in the stat obscures that.

    Vote -1 Vote +1

  8. StrikeThree says:

    Batter position in the box, forward or backward, allows a catcher (positioned a standard difference from the batter) to receive the ball at a higher or lower point in its trajectory, particularly higher with breaking balls, sinkers, split fingers low in strike zone. The height of the ball when received mayimpact umpire decisions. Batter position may vary according to count, pitcher, anticipated pitch, umpire

    Vote -1 Vote +1

  9. Alex says:

    Should this be broken down by count? We know that umpires are more likely to call a borderline strike in a 3-0 count or a 2-0 count than in a 0-2 count. Maybe the guys who are “bad” at getting strike calls are simply better at working the count in their favor, and then paying for it later?

    Vote -1 Vote +1

    • No Comment says:

      I was wondering this as well. How much of a poor Diff/1000 (more strikes than expected) for a hitter is the result of being in more hitter’s counts and how much of a good Diff/1000 (more balls) is the result of being in pitcher’s counts?

      Vote -1 Vote +1

  10. bampton says:

    I’d like to see a gif/screenshot of some of the balls from the first chart (Gonzalez?). There’s a few very questionable ones, and I see one that is nearly dead centre. Angel Hernandez could be involved in some capacity.

    Vote -1 Vote +1

  11. Jeremy says:

    What about the umpire’s position? Look at the two screenshots. First, are they the same ump? They seem to have the same number on their sleeve. Second, what was the resulting calls? The umpire for Ethier has his face about even with the batter’s side of the plate. The umpire for Gonzalez has his face on the opposite side. Gonzalez’s umpire might think the ball catches more of the plate from his position on the other side than Ethier’s. If there’s a correlation to the ump’s position, it comes back to the catcher’s positioning, doesn’t it? The umpire is trying to guard himself behind the catcher. If the catcher is set up away from the pitch, it looks like it might advantage the pitcher.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current day month ye@r *