Pitch Framing Is Evolving Along With the Strike Zone

Earlier this month, I wrote about the improvements that umpires have made in calling balls and strikes according to the rulebook strike zone. Today, I’d like to focus on the other side of that equation: pitch framing. The consensus around baseball is that pitch framing’s story has followed a very familiar arc. Call it the Competitive Advantage Life Cycle:

• Teams realize the immense value of a skill.
• An arms race ensues as they scramble to cultivate it.
• The skill becomes widespread across the league.
• Since the skill is more evenly distributed, it loses much of its value.

Once everybody got good at pitch framing, nobody was great at it. As Rob Arthur has put it, “Catcher framing felt like it was disappearing almost as soon as it was discovered.” I even have fun graphs to drive the point home. There are definitely more useful ways of presenting the data, but I like how these ones let you watch the entropy dissipate over time in open defiance of the second law of thermodynamics:

I spent entirely too much time giving each team its exact shade, which should really help so long as your favorite team’s main color is Miami Blue, Northwest Green, or Rockies Purple. Still, if you follow along at the very top, you can the see Braves starting out strong because of Brian McCann, then the Brewers enjoying Jonathan Lucroy’s golden years. Past 2014 or so, individual players have a harder time standing out, so the only teams that can break away from the pack are the ones with good framing duos: the Dodgers with Yasmani Grandal and Austin Barnes, the Brewers again with Omar Narváez and Manny Piña, then in 2021 and 2022, Jose Trevino with Jonah Heim in Texas and Kyle Higashioka in New York.

The compression is even more striking when you look at individual players. Trevino inhabited his own stratosphere in 2022, but even he couldn’t rack up framing runs the way the best framers did 15 years ago:

So that’s the story of pitch framing, and it’s been around for a while. Jeff Sullivan noted in 2016 that as the floor for framing talent rose, the standard deviation and year-over-year correlation in framing value started falling, and they have continued to fall. However, this narrative often leaves out a key piece of context. I would say that we need to reframe the conversation, but Zach Crizer beat me to that joke years ago. Catchers have definitely gotten better at framing, but that improvement hasn’t happened in a vacuum. It’s been concurrent with increased pressure on umpires to adhere to the strike zone as measured by PITCHf/x and Statcast. I think the two trends are more related than we realize, and I would love it if someone who’s smarter than I am could figure a way to analyze their effects separately. I tried all the math I know. At one point, and this is true, I told a friend that I was making box and whisker plots, but the friend had never heard of box and whisker plots, so they thought I was talking about some sort of cat cemetery. I don’t have concrete answers for you, but I did find some interesting things.

I should start by addressing the fact that umpires aren’t calling more strikes. How can everybody be getting better at pitch framing when umpires aren’t calling more strikes? It starts with improved plate discipline across the league. Batters are getting better at swinging at strikes and laying off pitches in the chase and waste zones. As a result, more of the pitches umpires have to call are outside the strike zone.

When the pitch tracking era started, catchers lost more strikes than they stole. That is, there were more pitches in the zone getting called for balls than there were pitches outside the zone getting called for strikes. Now let’s take a look at what happened when umpires started improving. Here’s a graph from my earlier article:

Overall accuracy has improved, and the gap has closed significantly. Combine that with the fact that most takes come on pitches outside the zone, and you end up with a stagnant called strike rate overall, but a rising rate in one particular area:

Umpires are calling more strikes in the shadow zone — that is, on the edges of the strike zone, where framing happens. In fact, in the shadow zone, they’re now more accurate at calling strikes than balls, which certainly lends credence to the notion that catchers have figured things out:

On pitches outside the zone, umpires are calling more balls, framing be damned. On pitches in the zone, the two sides are working in concert and the improvement is coming more quickly. (To be clear, I don’t think catchers are making this distinction; most pitches in the shadow zone are close enough that they’re worth framing.) The result is that catchers are stealing fewer strikes than ever, but they’re also losing fewer strikes than ever.

In 2009, Jose Molina led the league in strike stealing. On balls in the shadow zone, he fooled the umpire 39% of the time. However, Molina only converted 71% of would-be strikes in the shadow zone. That was enough to trounce the field in 2009, but it would have been the absolute worst rate in the league in 2022. As promised, here are box and whisker plots of called strike rate in the shadow zone, for pitches outside and inside the strike zone. No cats were harmed in the making of these graphs:

Bigger mistakes have largely disappeared as well. As I noted in my earlier article, umpires missed a call in the heart, chase, or waste zones just once every 2.74 games in 2022. There’s often an extenuating circumstance when an umpire calls a ball on a pitch over the heart of the plate. Strikes in the chase and waste zones are scarce as well. Brian McCann led the league with 300 of them in 2008. In 2022, Keibert Ruiz led the league with just 21. People have been saying for a long time that framing is less about tricking the umpire and more about presenting pitches well, but it’s never been more true.

As Jason Castro told reporters before the 2020 season, “My job is to keep strikes strikes. I mean, that’s really what I’m trying to do: Reward the pitcher as often as possible for doing his job. Everything outside of that is kind of like cherry on top, if you can get some extra pitches.”

These days, even the best framers can’t pull many strikes out of thin air. I ran a simple regression to predict pitch framing scores based on called strike rate in the shadow zone. Each year, as those stolen strikes dry up, the model weights pitches in the zone more and more heavily and pitches outside it less heavily. In 2010, pitches in the strike zone accounted for 73% of a player’s framing runs. In 2022, they accounted for 83%.

However, there are parts of the zone where catchers are both converting and stealing more strikes than they used to. Over the last few years, catchers have been working harder than ever to earn strikes at the bottom of the zone, and the one-knee catching stance has made its way around the league. It makes a certain amount of intuitive sense. The sides of the plate never move, whereas chests and knees change heights from batter to batter, and sometimes from pitch to pitch. It’s no surprise that catchers would focus on earning strikes in the squishiest parts of the zone. But it’s not that catchers are fooling umpires into calling more strikes at the top and bottom of the zone. Umpires get graded based on pitch tracking, and pitch tracking is telling them that those pitches are strikes. I pulled the called strike rate both inside and outside the strike zone for the top, bottom, and sides of the shadow zone, ignoring the corners so as to avoid overlap:

Umpires have grown more accurate in all three areas. On the sides, they’ve improved by calling more balls, but at the top and bottom they’ve done so by calling more strikes. In 2008, umpires almost never called a strike below the official strike zone. That’s no longer the case, and it’s made framing much more valuable at the bottom of the zone. At the top of the zone, umpires have gotten more accurate and consistent without changing the border. The average strike there crossed the plate just .03 inches higher in 2022 than it did in 2008. The sides are the same story. The average strike is actually .03 inches closer to the middle of the plate than it used to be on either side. However, at the bottom of the zone, the average strike has dropped by slightly more than an inch.

There is no such thing as a catcher who is good at framing in every quadrant of the zone. Baseball Savant’s framing leaderboard lets you see each catcher’s strike rate in each quadrant of the shadow zone going back to 2015. No catcher has ever been above average in all of them over the course of a season. There’s just too much plate to cover all of it well. However, the overall shift in the zone has opened up a new way to provide value. Catchers didn’t start crouching on one knee and focusing on the bottom of the zone by accident. There are more strikes than ever to be found there.

I don’t know how much longer this discussion will remain relevant, but as Patrick Dubuque noted recently, framing still provides plenty of value, and for now it should help us appreciate Jose Trevino while he’s still doing his thing. I imagine that if you go to FanGraphs at all, there wasn’t much information in this article that was completely new to you. All the same, as both pitch framing and umpiring have changed over time, it helps to understand the ways that they’re interdependent. Computers and cameras determine the dimensions of the zone, but the people tasked with enforcing those boundaries are human. Even as they get better at appeasing their robotic overlords, their humanity finds new places to peek through, and for other humans to exploit it.

Thank you to our resident catching expert Esteban Rivera, who emphasized today’s “keep strike strikes” focus and did his best to make sure I didn’t say anything egregiously wrong.