The Angels Have Had a Month of Clayton Kershaw

There are a lot of really good Clayton Kershaw fun facts, and he seems to generate countless more every start. He’s the best there is, and the best always produce the best statistics. I think my favorite Kershaw fun fact for now, though, is this: in 2013, he allowed a .521 OPS. In 2014, he allowed a .521 OPS. In 2015, he allowed a .521 OPS. This year, he’s somehow even better, and that doesn’t make any sense, but this year is still going. The fun fact captures the years completed, and it teaches you everything you need to know about Kershaw in a matter of seconds. He’s impossible, and he’s impossibly consistent.

Over the past few weeks, spanning five starts, Matt Shoemaker has allowed a .516 OPS. He’s walked one of 144 batters, while striking out 48. Now, Clayton Kershaw, Matt Shoemaker ain’t. Through the season’s first six starts, Shoemaker had allowed more than a run per inning. He was horrible! Now he’s Kershaw. He’s not really Kershaw, but, it makes you wonder, what’s the significance of looking like Kershaw for such a stretch of time?

I already went to the Shoemaker well two weeks ago. Alex Chamberlain went there last week. This isn’t a new subject, but then, it kind of really is. This is very quietly astonishing, if that’s even possible, and here’s a plot of some rolling averages. It’s been a dominant five starts, mostly. Here’s Shoemaker over five-game stretches:

Down goes the contact. Up goes the rate of swings at would-be balls. Promising trends, both. Shoemaker has been fooling batters with strikes and balls, and as we’ve written, he seems to owe much of this to a simple change in approach. The Angels say Shoemaker has been throwing a better fastball, and a better splitter. What’s easy to tell is he’s just been throwing a lot more of the splitter.

When Shoemaker was recalled from Triple-A to start on May 11, he threw 12% splitters. When he faced the Dodgers on May 16, he threw 43% splitters. Let’s cast aside the five-start stretch and examine the most recent six-start stretch, since that captures the whole of the splitter-heavy Shoemaker. The start against the Dodgers wasn’t great, but it was fine, which — for Shoemaker — was progress. Over the past month, Shoemaker has thrown 44% splitters. In second place: Nathan Eovaldi, at 28%. Jeremy Hellickson has thrown 28% changeups. Here’s how Shoemaker has ranked, among starters with at least 20 innings:

It’s convenient that our last-30-days split envelops Shoemaker’s six starts, and no more. There are 137 pitchers who have thrown at least 20 innings over the month. Shoemaker has been arguably the best of them. I mean, yeah, you’re right, Kershaw. But Shoemaker ranks in the top 5% in all statistics but ERA- and fastball rate. In fastball rate, he’s in the bottom 5%, and we all know ERA isn’t super reliable. Shoemaker has actually generated the lowest rate of contact allowed in the whole sample. He’s just barely in front of Kershaw and Jose Fernandez.

And in case you’re wondering, Shoemaker has been in the top 20% in batted-ball speed. Over the month, he’s allowed batted balls at an average of 87 miles per hour. Ditto Kershaw. Ditto Zack Greinke and Jake Arrieta. I didn’t know what I was going to find when I pulled up that leaderboard, but, here we are.

You understand pitching. You understand game theory, even if you think you don’t. It’s not just about throwing more splitters. It’s about Shoemaker throwing more of his best pitch, while making himself less predictable. He’s thrown a ton of splitters in fastball counts. He’s thrown fastballs in non-fastball counts. He throws a slider, too, which doesn’t make the task of hitting him any more pleasant. But by increasing the splitter usage, Shoemaker has probably helped his fastball the most. Before these six starts, opponents slugged .705 against Shoemaker’s fastball. His whiff rate on the pitch was in the bottom 20%. During the six starts, opponents have slugged .321 against Shoemaker’s fastball. His whiff rate on the pitch is in the top 3%.

Opponents have braced for the split, and then the fastball has been there. Or it hasn’t been, but they’re hard to tell apart. Early on, Shoemaker had five strikeouts on fastballs, and six strikeouts on splitters. That’s over those first six starts. In the last six, the numbers have been 17 and 28, respectively. Even in Shoemaker’s breakout 2014, the splitter was his putaway pitch. Now the fastball is helping, and you can see these pitches in action. Here’s Carlos Santana not expecting a 3-and-1 splitter:

Then a full-count fastball gets a wave:

This is the thing about a bad fastball. A bad fastball isn’t a bad fastball if you surround it with other weapons. Matt Shoemaker is a great example of how pitches work together. By getting opponents to worry about his splitter, everything is more successful.

Let’s jump back to a question posed early on. And now, for this, we’ll go back to focusing on the five-start stretch. Shoemaker’s walks: 0, 0, 0, 0, 1. Shoemaker’s strikeouts: 12, 11, 8, 6, 11. He hasn’t posted a Game Score under 50. Using the Play Index, I looked up stretches of starts with no more than one walk, and with at least six strikeouts, and with a Game Score of at least 50. This season, Kershaw stitched together nine such starts. Shoemaker is in second with an active streak of five. He’s tied with Fernandez, who also has an active streak of five. No one else is higher than three.

I ran the same search, going back to 2000. I wanted to see some of Shoemaker’s company. There have been 30 qualifying stretches of at least five starts. Some of the names: Kershaw, Fernandez, Roy Oswalt, Felix Hernandez, Jake Arrieta, Curt Schilling, and so on. The average seasonal ERA- of one of the pitchers with such a streak: 68. The average seasonal FIP- of one of the pitchers with such a streak: 66. Shoemaker is the only pitcher in the group who would have a triple-digit ERA-. Josh Tomlin is the only pitcher who would have a triple-digit FIP-. He came in at 110; second-worst is Mat Latos, at 85, in 2011.

The benefit of seeing the list of peers is you get an idea of how good you have to be in order to do something like this. Nearly all of the pitchers who have done what Shoemaker has done have been aces. It’s not mathematically impossible for a bad or mediocre pitcher to do this, but it’s far more likely the pitcher is good, or great. Great pitchers have a higher baseline, so higher levels of performance are more easily within their reach. Perhaps you’re hung up on seeing Tomlin’s name. I’ll point out his streak had 33 strikeouts, and eight home runs. Shoemaker is at 48 and two. Shoemaker’s streak blows Tomlin’s out of the water.

Given what Shoemaker has done, everyone wonders, so, how good is he? How much should we improve our mental projections, since even a month and a half ago he looked like a nightmare? Projections are inherently conservative, and Shoemaker hasn’t yet proven he’s a reliable ace. I think we all understand that. Yet for a month or so, Shoemaker has changed his approach and he’s pitched as well as Clayton Kershaw. You have to be a certain amount of good to spend a month pitching like Clayton Kershaw. Almost anyone in the majors could fluke into a no-hitter. It’s much more difficult to do what Shoemaker has, and for that reason, April is but a distant memory, as Shoemaker has rebuilt for himself a career.