The Ongoing Rise of the Pitch-Framing Floor

The Hardball Times Annual. You should buy it! The whole thing is good, probably. I’m admittedly biased because I wrote for the damned thing. My topic this time around was pitch-framing, and the article appears in the book, and in the book only. Everything in there is a book exclusive. That’s the way books work.

Now, here’s the deal: In the interest of maximizing sales, I’m probably not supposed to cover ground I already covered in my essay. But as you’ve presumably heard, the Twins signed Jason Castro Tuesday afternoon, and there’s an angle here I don’t want to let pass by ignored. So, it’s time for some overlap. In the book, I talk about all of this theory stuff in greater detail. But Castro going to Minnesota only further raises the floor of the worst pitch-framing teams.

Once upon a time, Castro was an interesting hitter. And he still is interesting, I suppose. There’s more ability in there than he’s shown. He projects for an 81 wRC+, but four years ago, he finished at 129. A 129 wRC+ for a catcher! Last year Manny Machado had a 129 wRC+. Castro’s offense has eroded, but his defense has improved. Eno talked with him about improving his pitch-framing some years ago. And improve, Castro has. It’s been the subject of most of the articles about the signing. Castro doesn’t hit so great anymore, but he rates well by the framing statistics. Doesn’t matter the source. He’s a good receiver, and it’s something the Twins evidently prioritized.

Why would the Twins in particular be so interested? Now to return to my simple strike-zone construct. I’ve mentioned probably dozens of times that I calculate expected strikes using data available here on FanGraphs. By looking at the difference between actual strikes and expected strikes, you can get a sense of who is and is not getting a favorable strike zone. Here is how the Twins, as a team, have rated over the PITCHf/x era, in terms of strikes above or below the average:

It’s pretty easy to tell from that when Joe Mauer started catching less, and Ryan Doumit started catching more. For five years in a row, the Twins have rated as being pretty strongly negative. In 2012, they ranked 27th in baseball. In 2013, they ranked last. In 2014, they ranked last. In 2015, they ranked 23rd. In 2016, they ranked 26th. The Twins haven’t been able to escape the bottom of the pack, and when you combine everything, it looks horrible.

One takeaway from that table: Way to go, Jonathan Lucroy! The Brewers have been way out in front, for reasons that’ve been exhaustively documented. the Twins are dead last, by hundreds of strikes. You can’t pin it all on the catchers. Maybe the pitchers have just been hard to receive. Most of them haven’t been very good. But catchers are responsible for most of this. Lousy receiving has cost the Twins so many runs, and they finally grew sick of it. So now they have Castro. He won’t make the Twins good by himself, but he’ll at least make the pitchers feel more comfortable on the mound. They’re going to get a little more wiggle room.

The Twins, now, have a good-receiving catcher. They got him from the Astros, but the Astros have already replaced him with Brian McCann, who’s a good-receiving catcher. McCann himself came from the Yankees, but the Yankees installed Gary Sanchez, who rates as a good-receiving catcher. You see the Rockies near the back of that list? In Tony Wolters, they finally have a good-receiving catcher. The Rangers have Lucroy. The Indians have Roberto Perez. On and on and on it goes.

It remains true that some teams get better strike zones than others. Some catchers remain better at receiving than others. But the floor has been steadily rising. The worst pitch-framing teams have by and large been getting better, closing the gap between the worst and the best. It’s easier to go from being bad to being okay than it is to go from being good to being great. The Twins could now further close the gap, but look at the evidence we already have through last season. Here are the year-to-year standard deviations of team extra strikes:

In 2011, the standard deviation was 219. Last year, the standard deviation was 108, or basically half what it had been. The spread between teams has been chopped by that much. As another way of visualizing this, here is a plot of just the difference between the best and the worst teams in each season:

In 2011, the Brewers were better than the Indians by almost a thousand strikes. This past season, the Cubs were better than the White Sox by 402 strikes. And there was also a big gap between the White Sox, in 30th, and the Tigers, in 29th. To this day, there can still be bad-receiving catchers and teams, but there’s less to be gained and less to be lost. The spread has been significantly reduced, and with the Twins adding Castro, just another bad-receiving team has elected to fix that. As the worst teams improve, the average improves, so half the teams still have to be worse than that, but below-average now doesn’t mean what it meant just a few seasons ago.

Under new management, the Twins are set to embrace the pitch-framing era. It’s long overdue, and just in isolation, it should be a great move for them. The pitchers, at least, will be happier. In context, it’s a little like hiring an analytically-oriented GM. It’s exciting to get a smart new executive, but it’s 2016. Almost everybody has an analytically-oriented GM. Differences are getting ever more narrow. I don’t know where this is going to lead, but the biggest inefficiencies are likely to be found somewhere else.