Clayton Kershaw got the win against the Mets last night. From one perspective, he was fantastic. He faced 25 batters, and struck 10 of them out, issuing but one single walk. He threw more than two-thirds of his pitches for strikes, and he pitched himself into the seventh inning. The problem was that, of the 14 batters who hit the ball fair, four of them went deep. Kershaw was charged with a season-high six runs allowed, giving him a game ERA of just about 9.
Let it be acknowledged right away: Kershaw’s still amazing. Kershaw’s still an ace. He still has one of the game’s highest strikeout rates, and he still seldom issues any walks. He’s near the top among starters in ERA-, and he’s even closer to the top by xFIP-. Kershaw might be literally the last pitcher on the planet you’d want to catch yourself worrying about. There’s just this one thing I can’t get past: Kershaw has coughed up 17 homers. That is, already, a new career high.
I know what you’re probably thinking. Lots of pitchers out there are going to set new career highs. Home-run spike, and everything. It’s absolutely true that the league-wide context has been changing, but this isn’t Kershaw getting caught up by a simple league adjustment. His home runs have gotten out of control.
For the first time, Kershaw’s home-run rate is worse than the league average. Perhaps it won’t remain this way, but we’re about halfway through the regular season, and Kershaw himself knows that he’s had an issue. I’ll excerpt:
“I’m giving up a lot of home runs right now,” Kershaw said. “You hope mistakes are hit for singles or doubles or whatever, it just so happens that mine are going out of the ballpark right now. I don’t know what you do to change that, other than stop making mistakes, I guess.”
Kershaw is right to not express too much concern. Of all the standard outcomes out there, home runs are the most noisy, and Kershaw knows he’s still executing most of the time. Of the fly balls he’s given up, 19% have left the yard; his career rate is a hair under 8%. To some extent, Kershaw has probably been bitten by bad luck, and this is going to regress. Regress in a good way, I mean.
I don’t think it’s all random, however. I think Kershaw himself is a little bit different. Here’s one supporting indication:
Kershaw’s contact rate is the highest it’s been since 2013. His zone rate is the lowest it’s been, ever. Fewer would-be strikes, more bats on balls. I don’t know if the theory I’m going to propose is the right one, but I won’t let that stop me. Let us think, first, about the changing baseball landscape.
I’m going to show you a plot I’ve updated since I first wrote about this a few months ago. This plot shows the rate of home runs per swing since the start of the pitch-tracking era in 2008. Using Baseball Savant, I looked at pitches thrown in the strike zone, split into three groups: low in the zone, middle of the zone, and high in the zone. Results:
As you know, baseball’s experiencing a home-run surge. It hasn’t affected all pitches equally. Home-run rates against high pitches are up, but they’re lower than they were in 2008-2009. So, there’s still a feeling of some normalcy. Home-run rates against low and middle pitches, though, have skyrocketed, relatively speaking. Middle pitches go for homers most often, and that’s what you’d expect, but also, compare the other two lines. Since the start of last season, low pitches have gone out more often than high pitches, on a per-swing basis. That’s not how things used to be.
The easy, over-simplified explanation: Pitchers have increasingly been working down. Hitters have worked to combat that by trying to elevate pitches down. Anything you’ve ever read about a hitter trying to hit more fly balls has involved that hitter trying to elevate pitches around the knees. Pitchers worked low, so hitters started to work low. Hitters got better lower in the zone. This doesn’t capture everything, but I do think it makes overall sense.
At last, we can return to Kershaw. What’s he been up to? Here are two heat maps, with pitches from 2014-2016 on the left, and pitches from 2017 on the right:
It’s possible it’s too hard to visually compare the two like that. I’d note that the bulk of pitches has moved downward, and there’s less activity around the zone’s upper edge. As a different way of examining the same data, here’s a plot of Kershaw’s rates of pitches thrown at least 2.5 feet off the ground. I used 2.5 feet because that’s approximately the average vertical middle of the strike zone.
Kershaw this year has thrown 30% of his pitches in the upper half, and 40% of his fastballs. Those are new lows — his previous lows were 39% and 51%, respectively. Compared to a year ago, Kershaw has dropped by double-digit percentage points, and we can break this down even further by isolating his primary pitch types. Here are Kershaw’s fastballs, sliders, and curveballs, examined in terms of their average vertical location when crossing the plate:
Kershaw’s average fastball has been lower than ever. His average slider has been lower than ever. At the same time, his average curveball has been up a little bit, relative to 2016. Overall, compared to a year ago, Kershaw’s average pitch has come in lower by 2.6 inches. Out of the 184 different pitchers I looked at, Kershaw’s got the eighth-biggest average drop. This isn’t just statistical noise — this is a real change in process.
Whether it’s accidental or deliberate, I don’t know. I definitely don’t know why Kershaw might’ve felt the need to change what he was doing. It was working for him, with high fastballs and occasional high sliders that functioned like cutters. It’s worth pointing out this is the first year in a while Kershaw hasn’t gotten to work with A.J. Ellis, but it’s not like he’s just been introduced to Yasmani Grandal, and it’s not like Kershaw couldn’t tell Grandal how he likes to pitch. He’s Clayton Kershaw. He can do whatever he likes. He’s started mixing in pitches from a lower arm angle, just because. Kershaw gets the benefit of the doubt, but as of right now, in 2017, he’s been working lower and lower, at a time when hitters are looking lower and lower.
That could at least help to explain why Kershaw’s numbers are where they are. He’s simply been pitching in a more homer-prone way. And by pitching up less often, Kershaw’s allowed hitters to increasingly target a smaller area. Right-handed hitters can look for something around the low-inside quadrant. And I should also point out that lefties have taken him deep five times, after failing to do so at any point in 2016. So this isn’t just righties by themselves. Kershaw’s locations have changed, and so have his results.
It doesn’t make the challenge of facing Kershaw all that much less challenging. He is, again, still very much a No. 1, and he’s more than earned a measure of patience. For all I know, next time out, Kershaw could return to the upper half with a vengeance, and that could be that. But for this year’s 15 starts, Kershaw’s thrown lower. That’s not how he’s accustomed to pitching, and it’s not how opponents are accustomed to seeing him. It would be nice to know why the change has taken place. The best I can say is that the change has happened.
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.