Testing Dexter Fowler and the Hitter Strike Zone
Just because of the sheer number of taken pitches in a season, hitters will frequently disagree on a call with the judge standing behind them. A certain number of disagreements are to be expected, but from the sounds of things, Dexter Fowler feels like he’s been disagreeing more than usual. Fowler doesn’t think his approach is worse, but his numbers are worse, so he identifies some iffy calls as a factor. This is what Fowler looks like after an iffy strike call:
If that isn’t enough, here’s another example:
If that isn’t enough, here’s another example:
Late last week, Fowler was frustrated after a game in which he struck out looking three times. One of those came on a certain third strike, but here are the other two:
Both times, the third strike was at least questionable. Both times, it came in a full count. It’s easy to understand why Fowler would be upset, given what he and his Chicago Cubs are playing for. This also presents a good opportunity for investigation. There’s a theory, thus, that Fowler has kind of gotten screwed this year, compared to the average. Might it be true? And in the bigger picture, who has gotten screwed, and who’s hasn’t, from the hitter’s side?
It’s time to pull out a familiar tool: the difference between actual strikes and expected strikes. I usually do this when I’m looking at pitchers and catchers and whatnot, but it can also apply to hitters. Here’s an example of something I wrote on the matter a couple weeks ago. We know how many strikes there have actually been. With zone rate and out-of-zone swing rate, we can calculate an expected-strikes number. Then we just find the difference. It’s not perfect, but it is quick and easy, and it can tell us something about who has and hasn’t had a strict called strike zone.
For this year, I looked at guys who’ve faced at least 1,000 pitches. Here are a couple top-10 lists. First, the biggest positive differences between actual strikes and expected strikes. Then the reverse. The former means the hitter could be said to have gotten screwed. The latter implies the zone has been tight, and therefore hitter-friendly.
More actual strikes than expected strikes
- A.J. Pollock, +33 strikes
- Alexei Ramirez, +31
- Trevor Plouffe, +29
- Seth Smith, +28
- Dexter Fowler, +28
- Eric Sogard, +26
- Wilmer Flores, +25
- Mark Reynolds, +25
- Yunel Escobar, +24
- Ian Kinsler, +24
Based on this quick glance, Pollock has had the most pitcher-friendly zone. Perhaps more notably, there’s Fowler, right there in fifth, a fraction of a strike behind Smith. So at this point, Fowler’s theory appears to be somewhat valid. We’ll come back to that. But first, the other list:
Fewer actual strikes than expected strikes
- Adrian Gonzalez, -41 strikes
- Mike Trout, -35
- Prince Fielder, -28
- Nick Markakis, -28
- Brandon Moss, -27
- Dustin Pedroia, -26
- Manny Machado, -26
- Evan Longoria, -25
- Bryce Harper, -24
- Jason Kipnis, -24
The opposite of A.J. Pollock would be Adrian Gonzalez, who’s so far seen a very friendly strike zone. There’s a lot of talent on this list, so there could be a relationship there. Perhaps it’s umpires giving certain hitters the benefit of the doubt. Perhaps it’s these players knowing which pitches are the right ones to lay off. Perhaps it’s something else, or randomness. Whatever the case, these called balls aren’t hurting. Against most of these guys, pitchers need every strike they can get.
Let’s return, somewhat, to Fowler. This season, in terms of the difference between actual strikes and expected strikes, he ranks in the sixth percentile. That’s obviously bad for him, but Fowler’s complaint is more that things have been worse for him than normal. Is it possible this is just his normal? For comparison: A season ago, Fowler ranked in the 28th percentile in the same statistic. Still not good for him, but definitely better. So by that indication, Fowler’s zone has indeed been worse.
We can take last year’s rate and apply it to this year’s numbers. As noted, Fowler right now is at +28 extra strikes. After this many pitches, by last year’s rate, he would’ve been at +9 extra strikes. That gives him a year-to-year increase of about +19 strikes. That increase ranks 10th-highest in baseball, among hitters who’ve seen at least 1,000 pitches. Didi Gregorius comes out at No. 1. He’s followed by Robinson Cano and Reynolds. These are guys whose strike zones appear to have gotten more difficult (for them).
At the other end, Markakis has the biggest drop. Then Machado, then Harper, then Adam Lind. These guys have gotten more friendly strike zones. Relative to a year ago, they’re seeing fewer called strikes, and more called balls, adjusted for changes in zone rate and so forth.
Leaving the microscope over Fowler just a little bit longer, this year, by Baseball Savant, he’s tied for second in baseball in called third strikes out of the zone. He has 19; he had 13 last year. Meanwhile, StatCorner offers some useful statistics. There’s the rate of pitches taken in the zone called balls, and there’s the rate of pitches taken out of the zone called strikes. Fowler’s changes:
Pitches taken in the zone called balls
- 2014: 87th percentile
- 2015: 72nd
Pitches taken out of the zone called strikes
- 2014: 53rd percentile
- 2015: 16th
What the first one addresses is a drop in rate, where Fowler this year has gotten fewer called balls in the strike zone. The second one addresses a hike in rate, where Fowler this year has gotten more called strikes out of the zone. The entire landscape agrees with itself; from the indications, Dexter Fowler this year has dealt with a less-friendly strike zone than he did last year. His shift has been relatively dramatic. It’s easy enough to dismiss hitters complaining when they’re frustrated, but Fowler’s eyes haven’t lied. He’s not the only guy getting hurt, but he doesn’t need to care about the other guys.
Hitter strike zones are less of a skill than pitcher/catcher strike zones. Hitters have only a little to do with the zone that gets called, so for any number you see above, you should mentally be regressing pretty strongly toward the league average. But every call does count, and in certain cases, you can kind of understand a pattern. Fowler, for example, shares a division with the Cardinals, Pirates and Brewers, who have pretty well established pitch-framers. That’s going to have an effect, and it’s going to be frustrating. In the end, some things will even out. You’ll have to forgive Dexter Fowler and a handful of other players if they’re skeptical.
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.
If I dont swing, that means it was a ball. Wanna fight about it?
Challenge accepted.
Just out of curiosity, did Gregory Polanco qualify to be in the list? Pirate fans are pretty sure that he’s getting screwed on ball/strike calls big time.
If I added up his TexasLeagurs page, he’s faced 958 pitches, so he just falls short of qualifying.
…I feel kind of dopey, though, because I can’t figure out how to do the calculation.
Does pitch f/x track the batter’s location relative to the strike zone? It would be interesting to see if that was at all related to how often umpires “miss” calls.
I also wonder if there might be some relationship between wRC in previous years and strike zone luck – are known good hitters benefiting more often?
I have nothing to add, statistically, to support your thought, but I umpire high school baseball games and the location of the batter definitely impacts the strike zone I call.
When a hitter crowds the plate against same handed pitching, it is really tough for the umpire to see the ball out of the pitchers hand, so we are basically calling the pitch as it is caught, with no anticipation based on the trajectory of the pitch.
Also, crowding the plate takes away any pitches that might otherwise be called strikes on the inside corner. It is very tough to call a strike on a pitch that almost hit a batter. Meanwhile, if the batter stands off the plate quite a bit, we can call that inside pitch more often and get some pitches off the plate inside as well.
If a batter check swings at a pitch, or shows bunt and then pulls back, it is going to “feel” like a ball. THat doesnt mean it will always be called a ball, because sometimes that pitch is right down the middle, but on a borderline pitch, the hitter will usually receive a ball call if he checks his swing. This could even extend to how a player tracks the ball into the zone. If the batter watches the pitch closely and leans toward it (on an outside pitch) or away from it (on an inside pitch), it definitely makes the pitch “feel” like a ball more often.
Now, I am not a professional umpire and I am not judged by any kind of pitch f/x system….but that is just my two cents to add to this conversation.
Although I do not fault you or any other umpire, this human, and perfectly justifiable response, bothers me because we have tools to eliminate it. The strike zone is a finite area of space that has been defined and redefined as evidence by the link below.
Yes, the player’s height and stance can adjust the vertical dimension, but the horizontal dimension is fixed because home plate has a fixed width. The idea that a strike zone can differ from player to player based on position in the batter box really bothers me.
http://mlb.mlb.com/mlb/official_info/umpires/strike_zone.jsp
I want to stress that I am only a high school level umpire and we do NOT have the tools to eliminate it. lol
Pitch f/x exists in MLB/MiLB baseball only. Personally, I would love to have the strikezone called for me. It would make my job a hell of a lot easier (although, I would likely be paid less….so maybe I would rather not see that happen).
But I agree that the MLB strike zone needs to be automated. No good reason why we shouldnt plug a microphone in the plate umpires ear and tell him what to call.
What I was trying to describe is the difficulty of human eyes to call the strike zone accurately when the variables surrounding the situation are constantly changing (The catcher, pitcher, hitter..etc.). I think with our current system, there are thinigs batters can do to shrink the zone that is called against them.
For a demonstration of the inside pitch effect, Anthony Rizzo famously moved closer to the plate at the beginning of the 2014 season. Compare his called strikes on the inside column of the strike zone since 2014 to his called strikes before (there are links to heatmaps there).
Who are the best hitters that have had the worst strike zone luck, and the worst hitters who have had the best strike zone luck?
I only got a few hours of sleep last night, and read this as “the Hitler Strike Zone.” Today calls for a nap.
The Hitler Strike Zone?
London, right?
Would be interesting to look at this data presented together with the data regarding the various counts the batters are in most frequently. It would seem that batters with a relatively high amount of 3-0 counts will have borderline pitches called strikes more often than batters who end up in a lot of 0-2, 1-2 counts. I doubt it is nearly this simple, however, the impact of the count on the way a pitch is called may be a significant factor.
It seems fairly intuitive, also, that batters who swing more are going to have fewer opportunities to get screwed by the umpire (obviously). I am assuming that the methodology for determining expected strikes is controlling for swing rates, and is only looking at called strikes, correct?
Granting these two points and taking them together, it suggests that the profile of a batters swing rates in various counts, relative to other batters, could be a factor.
Yeah, practically speaking I think you really have to factor in the count. We know that’s a pretty big effect… Or does Jeff already so this?
Actually, huh, the leaderboards surprise me. The “favored” list has some guys with good command of the zone, who would be in hitters’ counts, and tend to get extra strikes. Which is the opposite.
More than just how often a player gets into specific counts, we need to look closer at how often a player swings in specific counts.
It is fairly well understood that batting averages are higher generally in favorable counts, but it would be interesting to look at swing % in different counts, as well as in-vs-out of zone swing % in various counts. We talk about “patient” and “aggressive” hitters all the time, but there are tremendous differences in approach in different situations.
I was going to say something of the sort. I wonder if players who tend leave parts of the border strike/ball zones alone will get better calls from umpires than guys who might take more pitches, but tend to swing at borderline or off-the-plate pitches anyway.
Mark Reynolds and Adrian Gonzalez are good examples. Both can take a good amount of pitches. But Reynolds tend to get beat on stuff down and in and away (inside mostly off the plate). Outside pitches, especially on full counts, might go the pitchers way since he doesn’t present anything for the umpire to assume his case.
On the other hand, Gonzalez rarely throws his bat on outside pitches, and hammers inside pitches. Pitchers will try to hit that corner, and a good amount of times the umpire will give him the benefit of the doubt over the pitcher since he has A) let the ball pass enough times for umpire to see) and B) never swings at that pitch to demonstrate to the umpire it may not be in the zone.
I’ve seen the same thing with him when he played for the Sox, as well as players like JD Drew and Ortiz, who I’ve seen in recent memory get a lot of outside corner strikes called balls (if you could look him up in past data and see if I’m on to something here).
This article from a year ago hints that maybe Fowler’s rightfully steamed.
thanks an interesting article
“Both times, the third strike was at least questionable. Both times, it came in a full count.”
Both times there was also a VERY similar pitch earlier in the at bat that was called a ball.
The **bway referee theory. I’ve talked about this a lot in the past. I don’t mean to reference the video but its funny.
https://www.youtube.com/watch?v=wiGgoFMSr-A
He plays numerous games versus St. Louis. In your study, you need to factor in the whine and cry about every close pitch effect.
I have no statistical evidence to back this up, but could Dexter be having trouble because 3 of the best framers in the sport reside in his division? (Lucroy, Yadier, and Cervelli)
Speaking of Anthony Rizzo, where would he fall on this chart? I figured he would be somewhere around Adrian G.
nah he gets a hbp on strikes instead of a ball.
Where does Jorge Soler rank on this list? He may not have enough plate appearances to qualify but it would seem he’s being jobbed also!