Statcast’s New Catcher Throwing Metric Is Here
What makes you happy? Among the things that bring a smile to my face, talking about catching is up there. I will look for any excuse to write, talk, or think about catcher defense. I’m one of those people that has missed catching bullpens since I’ve stopped playing regularly. For those of you who do not know, that is not normal! So whenever Statcast drops new information about catcher defense, I have to write about it.
A few weeks back, I covered some catchers who were throwing at a rate that suggested additional defensive value relative to their peers in the new, more aggressive stolen base environment. Soon thereafter, MLB.com’s Mike Petriello revealed a new stat, Caught Stealing Above Average, to the public, and that Baseball Savant would roll out a leaderboard that would offer a more objective look at throwing out runners relative to the traditional caught stealing stat. You can check out the full leaderboard here.
There are multiple components taken into consideration for Statcast’s model that try to even the playing field when it comes to throwing out runners — variables like pitcher delivery speed, a runner’s lead and jump, and more. Evening all of those out provides more insight on how some catchers are more deserving of outs than others. Typically, I would highlight the catchers who have excelled at throwing out runners, but to emphasize the value of this statistic, I instead want to look at those who have been unlucky this year and last despite consistent strong throws, as well as other catchers where the trends are concerning. The first of this group is expected but notable nonetheless:
J.T. Realmuto (1 Catcher Caught Stealing Above Average in 2023, no. 7)
Base stealers have been running like wild against the Phillies this season, and it’s made for some confusing statistics for J.T. Realmuto. Out of his first 17 stolen base attempts of second base, he’s only caught five runners. If you remember this piece back in November, you know Realmuto has one of the strongest arms and fastest pop times in the game; if anybody should be throwing out most runners, it’d be him. But this year, he is only running a 29% caught-stealing rate, partially due to plays like the one above, where his pitcher was just slightly too slow to home. He has still been better than his expected rate of 22%, but his bar is much higher than any catcher in the league.
Realmuto’s 2022 track record is even more impressive than this year’s. If you combine all of last season’s attempts with this year, his CS% sits at 48%, with an estimated CS% of 23%. His laser-quick pop time makes up for his slower-to-the-plate pitchers. Realmuto is elite at throwing, framing, and blocking; nobody else can make that claim to this extent. We are watching one of the best defensive catchers of his generation.
Shea Langeliers (2 Catcher Caught Stealing Above Average in 2023, no. 5)
Not many things are going right for the A’s, but Shea Langeliers has impressed on both sides of the ball. With a 109 wRC+ fueled by a .244 ISO, he is off to a solid start with the bat, and both his swing and power are promising. On the defensive side of the ball, he has averaged a 1.95 pop time but has been a little unlucky with throwing out runners, with a CS% in his first 19 attempts to second base of 32% but an estimated mark of 24% — a similar discrepancy as Realmuto. On this throw, he was as perfect as you can be, but his pitcher was dragging his feet to get the ball home, and Tony Kemp lost his glove on the tag anyways.
While the pop time isn’t elite on average, Langeliers has proven that he is accurate enough to throw runners out if given the opportunity. It’s one of those situations where he isn’t necessarily a top tier thrower and therefore doesn’t have much room for error if his pitchers don’t cut him some slack or if his throw is just slightly off line. To me, Langeliers is the type of catcher who offers a glimpse into the future. With some automated ball-strike system inevitably coming to MLB, his profile is one that would perfectly transition to that new reality, as he is only average at blocking and framing but is slightly above average at throwing. Combine that with above-league-average offense, and you have yourself a perfect prototype for the potential new catching environment.
Martín Maldonado (2 Catcher Caught Stealing Above Average in 2023, no. 4)
If it weren’t for Realmuto, Martín Maldonado would find himself atop the new throwing leaderboard from 2022 through today, with seven CS Above Average. This year is no different, as he holds a 33% CS% despite a mere 14% estimated rate. The pitchers in Houston aren’t helping him, but he is nabbing runners regardless. Case in point: the play above, where Cristian Javier’s big leg kick and loopy arm took up too much time, undermining Maldonado’s pinpoint throw. But even when he isn’t catching runners, he is making it close.
Over the past few years, I’ve gone back and forth on the Astros’ decision to use Maldonado as their primary catcher. He is an incredible fielder but is consistently one of the worst hitters in the game. As his framing skills have undergone a slight regression compared to earlier in his career, it’s statistically unclear if it’s worth keeping him as the mainstay. But the additional data on Maldonado’s elite throwing and blocking in the last few months reminds us why he’s the lead catcher in Houston. Combine that with his glorified intangibles, and you can easily understand why he has cemented himself as the starter despite the lack of offense.
This new information can work in the other direction, too. Sometimes catchers are unlucky because their pitchers put them in a hole, and other times they can be even worse than expected. Unfortunately, there are some promising young catchers who fall into the latter category. Keibert Ruiz is 2-for-17 (12%) to start the year despite a 23% estimated CS%. His plus-2.00 pop time is the main reason for this. If his framing continues to trend in the wrong direction along with his arm, he will need to be an above-average hitter to live up to his prospect pedigree. Francisco Álvarez faces a similar dilemma, going 0-for-12 to start the year. His estimated CS% was only 12%, but his throws have not been competitive regardless.
For a while, catching was a semi-mystery; we knew who had rocket arms and who didn’t. Now, we have information about framing, blocking, and throwing that helps us figure out the true value of a gifted defensive catcher. It’s an exciting time to be a catching fanatic.
Esteban is a contributing writer at FanGraphs. One of his main hobbies is taking dry hacks every time he sees a bat.
this is awesome! data going back to 2016? my morning is now all booked up…
realmuto is just so much better than everyone else at this, it’s bonkers. he is at 41 CSAA and the next 3 are at 27, 26, and 25. 10th best is at 12. he also has the best pop time and 2nd best arm strength. oh, and he hits a lot too.
Some of this is because he’s elite at everything, and a lot of this is that he plays way more than everyone else (which is quite an accomplishment if you think about it, he’s good enough to hit every day and catching everyday is not easy on the body).
But the real outlier (and probably the real separator) is the pop time. Realmuto has always had excellent pop times but in recent years his pop times have gone from elite to nearly impossibly good. The best pop times for a catcher typically top out near 1.90. That was Realmuto for a long time, with pop times of 1.89 or 1.90 from 2017-2019. And then in 2020 he just leveled up; he’s now posting pop times closer to 1.80 than 1.90. His pop times in 2020 and 2021 were 1.83; last year it was 1.82; this year it has been 1.81.
This is so far beyond anyone else it’s not funny. There are other guys out there with stronger arms, like Jorge Alfaro and Gary Sanchez. Sean Murphy and Martin Maldonado have comparable arms to Realmuto. His transfer time is elite, but not an outlier (he’s in the same range as Christian Vazquez, Salvador Perez, and Tucker Barnhart). Nobody is even close to Realmuto in pop time.
This metric (and entire leaderboard, really) also helps explain a bit why Elias Diaz still has a job. Although only a bit.
(And for some reason he’s also inexplicably the best hitting player the Rockies have.)
Diaz is probably the single player who has the most to gain from an automated strike zone, since he’s so bad at framing and he’s so good at throwing out runners. Willson Contreras was my previous best example, and the leaderboard backs that up too. Langeliers was mentioned in the article, and I would tend to agree with that.
When one has watched baseball since 1948 it is rare when he is absolutely blown away by something that, despite being right before his eyes, he couldn’t see but that is case with Conner Wong. Forget that he is presently doing an unbelievable imitation of Johnny Bench with the stick, it is that he is also doing an actual Johnny Bench type performance with the mitt and the arm. He has already earned a full 1.0 bWAR on defense, Realmuto, great player that he certainly is, has only a 0.2 bWAR, but no one needs any system to reinforce the obvious. Is he really this good? It wasn’t easy to evaluate him in the sporadic appearances he had made before this year but when seen on a daily basis his prowess just jumped out. I will be watching with eager anticipation to see if he continues to amaze.
On Maldonaldo, I can see why the Astros used him prior to 2021. The throwing does not make up for the hitting and the decrease in framing numbers. 2021…Sure, that could have just been a down year. Since 2021, 0.2 fWAR. It is not easy to understand why he’s still the primary catcher.
He’s essentially Linus’s security blanket for the Astros pitchers. It appears Dusty Baker thinks he’s still playable most days. Maybe Dana Brown can have better luck in changing Dusty’s mind than Click did in the regular season.
As we keep getting better at quantifying aspects of catcher defense with framing, blocking, and throwing. Maldonado represents the last frontier that we haven’t quantified yet, staff management and game calling. The Astros clearly think he is uniquely excellent at those aspects (I recall Kevin Goldstein mentioning it several times) and that the value he provides there compensates for his other shortcomings.
I really would love to know if the Astros have found a way to quantify that and if so how big the scale of the effects are. I was surprised by the range of framing value when that first became public, staff management could be the next frontier if that can ever be measured confidently.
I expect this will be Maldy’s last season in Houston as I don’t think the Astros FO still agrees that these intangible qualities are making up for the easier to measure shortcomings. The Astros FO have tried to acquire Contreras a couple of times since last trade deadline to replace Maldy as the primary catcher. Considering Contreras is considered poor in that department, the upper bound variance between a great game caller/staff manager and a poor one is about 25 runs (but probably less).
I wish we had data for when Yadier Molina was in his prime.
I was just thinking that while reading the article.
Cool article. The little I’ve seen of William Contreras behind the plate this year has been impressive. He’s shown improvements from last year’s numbers and I wonder if you could do an article on Walker McKinven/Brewers on their reputation for developing catchers and what they do different.
They turned Omar Narvaez into a good defensive catcher, which seemed totally impossible.
Given that, what they did with Contreras isn’t that shocking. Contreras was kind of a mess in his rookie year but that’s very different from Narvaez being almost impossibly bad for two separate teams over multiple years. But it is super impressive that they appear to have dramatically improved the defense of two catchers now.
It’s not exactly a isolated Catcher stat. But what if you took SB allowed + WP + PB – CS + Pick Offs? Would that be valuable in managing baserunning?
These are catcher pickoffs, right, so shouldn’t they be subtracted rather than added?
Wild Pitches should be weighted much less than Passed Balls, as they are primarily the pitcher’s fault. Some blame should still go to the catcher, but not nearly as much as Passed Balls do.
Otherwise, this looks pretty good as a basic metric. I especially like how you’ve essentially used net stolen bases allowed rather than a percentage to help account for how the better throwing catchers dicourage stolen base attempts made against them in the first place.