Clayton Kershaw’s Replacing Strikeouts with Strikeouts, Basically
Clayton Kershaw’s good! Here’s something I bet you didn’t know about him. In the first half of this season, he struck out more than a third of all the hitters he faced. In the second half, his strikeout rate is actually down 17%. Now, that’s percent, not percentage points, but it means one of six strikeout victims hasn’t been a strikeout victim. That seems like the kind of thing that should raise eyebrows. But you haven’t noticed because in the first half Kershaw allowed 19 runs, and in the second half he’s allowed 19 runs. One is less inclined to notice when great players are slightly differently great.
Also, his second-half strikeout rate is still extraordinary. Also, he’s still not really walking anybody, even though just yesterday he did put Yusmeiro Petit on base. The regular numbers love second-half Kershaw, but if you dig just a little bit deeper, you can gain a better understanding of how Kershaw has remained so dominant despite giving away a handful of whiffs.
Here’s a clue. From Sunday, Kershaw vs. Andrew Susac:
From the same game, Kershaw vs. Matt Duffy:
You’re looking at a couple of pop-ups. Now, granted, Duffy’s pop-up dropped in between fielders, allowing it to be a run-scoring hit. For Kershaw, that was a bad outcome, but it’s an extremely rare outcome, and all things considered he would’ve been happy with the batted ball. This season, on pop-ups, hitters are batting .021, and slugging .025. They’re almost automatic outs. In the NFL, extra-point success rate hovers around 99%, and more and more people wonder whether the kick should even be necessary. In baseball, a pop-up is a lot like an extra point. The same thing doesn’t always happen, but when it doesn’t happen, it’s highly unusual.
So you know something’s going on with Kershaw and pop-ups. Why don’t we take a look at first- and second-half splits?
| Half | K% | BB% | HR/FB% | BABIP | IFFB% | ERA- |
|---|---|---|---|---|---|---|
| 1st | 34% | 4% | 8% | 0.279 | 5% | 50 |
| 2nd | 28% | 5% | 6% | 0.252 | 24% | 46 |
In the first half, Kershaw’s numbers were absurd. In the second half, he’s changed things up a little bit. He’s allowed more contact, but he’s also allowed worse contact, and that pop-up rate is the highest in the major leagues. His first-half pop-up rate was among the league’s lowest. Earlier, Clayton Kershaw was piling up the strikeouts. He’s still getting plenty of strikeouts, but he’s replaced some of them with the ball-in-play equivalent of strikeouts.
By the numbers available on FanGraphs, Kershaw generated three first-half pop-ups — two on May 23, and one on June 13. So, pop-ups accounted for 1.3% of his batted balls. In the second half, he’s had just one start in which he didn’t generate a pop-up, picking up a total of 16 overall. They’ve accounted for 7.5% of his batted balls. That tells you a lot, but we also draw from a conservative data source, and Gameday provides its own classifications.
According to MLB, first-half Kershaw generated ten pop-ups. Second-half Kershaw has generated 31. And, mostly, this has been about righties. Kershaw’s pop-up total against lefties has risen from four to five. His pop-up total against righties has risen from six to 26. That’s an extremely high total, and, where have those pitches been going? You could guess, but we can confirm with the help of Baseball Savant:
Mostly, the pop-ups have come against pitches inside and elevated. And if you examine Kershaw’s pitch patterns, you can see that perhaps this is deliberate. Kershaw has moved more often toward that quadrant.
Looking at his overall pitches against righties:
First half: 22% inside, up
Second half: 29%
Looking at his fastballs, still against righties:
First half: 31% inside, up
Second half: 40%
And how about if we look at first pitches? We’ll look at this two different ways:
First half: 70% inside
Second half: 86%
First half: 27% inside, up
Second half: 45%
More recently, Kershaw has thrown more first-pitch fastballs, and he’s thrown more of them high and tight. In the second half alone, Kershaw has generated 12 first-pitch pop-ups, 11 against righties. The second-place numbers are seven and five, respectively. Kershaw, on the season, has been pounding the zone to open at-bats. Batters, then, want to be aggressive early, because they don’t want to fall behind in the count and have to try to put a bat to Kershaw’s breaking stuff. So with batters looking to swing at 0-0, Kershaw’s moved a little closer to the hands, and more batters have gotten themselves out in a hurry. It’s a somewhat infrequent thing, but all of them add up.
There have been a few elevated breaking balls. Kershaw’s command of his slider is such that he can throw the pitch up and in from time to time, and over the past couple months, Kershaw’s slider has generated seven right-handed pop-ups, against one in the first half. But this is mostly about his fastball — that pop-up total’s risen from four to 17, looking at only right-handed batters. This is what you’d expect. The high, inside fastball is the pop-up pitch, and Kershaw’s thrown more of it. His fastball is also one of those fastballs hitters tend to swing underneath, because of its vertical movement. Kershaw has a long track record of generating pop-ups. They started to go away a little bit when Kershaw slightly altered his approach, but now they’re back with a vengeance, as Kershaw now is blending everything. He’s a strike-thrower like never before. He’s still not allowing much contact. Now he’s also limiting the quality of contact. Kershaw’s simply turned everything up to 11, blending his new skills with his old ones in becoming the perfect pitcher.
So in the first half, Kershaw struck out 34% of batters. In the second half, he’s struck out 28% of batters. But, Duffy’s pop-up aside, let’s go ahead and consider pop-ups automatic outs. That much isn’t always true, but then strikeouts aren’t always automatic outs, themselves. Blending strikeouts and pop-ups, first-half Kershaw turned 37% of batters into easy outs. Second-half Kershaw’s up at 38%. In one sense, the last couple months, Clayton Kershaw’s allowed more contact than he did before. In the truest sense, batters have just found more efficient ways to whiff.
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.



“on pop-ups, hitters are batting .021, and slugging .025”
This is totally unrelated to the article, but I want to see the infield popup that went for a double.
There are several of them! I’m guessing a lot happened where an outfielder dove and missed.
That outfield is a weird collection of gifted athletes, high salaries and somewhat underwhelming performance even though they are still quite good. I just don’t know how else to describe them.
And Sam Holbrook wasn’t umpiring.
I think those are league-wide numbers, not specifically the Dodgers’ allowed line on pop-ups.
My bad, you are correct.
I suspect this is more the batters doing than Kershaw’s. Batters are trying not to strike out against him so there are more good outcomes, the pop up that lands for a hit.
I’m not sure I buy that, especially in light of the first-pitch numbers cited above. Looks more to me like an All-World player adjusting mid-season to stay one step ahead of his competition, and flexing his muscles while doing so.
You mean Kershaw isn’t pitching differently? Because Jeff explains how Kershaw is pitching differently.
I also think it’s more about the hitters switching their approach than Kershaw pitching any differently. Once word got out that Kershaw had transcended to a higher level of pitching, the hitters aren’t waiting around taking as many pitches. You basically have to guess fastball with Kershaw and hope to get lucky, because you aren’t touching his secondary pitches once he gets ahead in the count.
As it happens, in the first half, righties swung at 41% of Kershaw’s first-pitch fastballs. In the second half, they’ve swung at 41% of Kershaw’s first-pitch fastballs. (I’m also a little surprised, but, here we are.)
Not related to what this post is specifically addressing, but I have noticed that it seems like sometimes Kershaw will start a game racking up a ton of strikeouts, but sometimes making a lot of pitches. Then, after K’ing 6 or 7 guys through 4 innings, he’ll strikeout 2 or 3 through the next 4 innings, but make far fewer pitches. It’s like when he’s on, sometimes he’ll start a game and the batters will try to be patient, and he’ll just strike them all out. Then, they’ll realize if the don’t swing, they get behind in the count so they’ll start to swing early, which just means Kershaw can cruise through the rest of the game on fewer pitches.
Interesting thought, and my own recollections track similarly… but the data doesn’t. He stays pretty consistent throughout the game: From BB-ref, Kershaw’s K-rate by inning:
Split K%
1st inning 32.6%
2nd inning 31.3%
3rd inning 32.2%
4th inning 27.7%
5th inning 39.3%
6th inning 30.4%
7th inning 26.1%
8th inning 35.6%
9th inning 33.3%
Of course, Kershaw can’t keep up with his first-half K%. Regression, hellllooo? A pitcher could pitch exactly the same way and face different outcomes. He could have the same approach, but have his pitches hit different zones, because no pitcher actually have pinpoint control. He also would face different hitters. Different stadiums, different weather, etc.
Sorry for being a bit condescending, but I’m honestly surprised that Fangraphs would publish such poor analysis.
I think the point was that at-bats that would have gone to strikeouts have become pop-ups, which are for all intensive purposes just as good as strikeouts.
Not bad analysis by any means.
It’s “all intents and purposes,” unless you’re referring to purposes of, relating to, or characterized by intensity.
Sometimes the purpose of a strikeout is to be very intense.
Grasping at straws here…
I’m guessing most people at fangraphs and a good part of their readers know that as well. Still it highlights just how good he was in the first half and how good he has been since his “regression to the mean.”
I don’t think simple error in command can create the dramatic shift in approach that Jeff observes here. There’s really no way that Kershaw’s rate of first pitches inside and up has increased from 27% to 45% between the two halves by chance alone. Pitching approach, like batting approach stabilizes relatively quickly.
I wish I could edit my comment to make it less assholish. Nothing wrong with speculating why a pitcher’s numbers had changed, but IMO such articles should start with acknowledging the most likely cause. Yeah, regression and random variance.
I’d much rather read an article that shows you an actual change in a pitcher’s approach and results that have coincided with that effect than one that rambles on and on about how every change we see in every player is probably regression. At this point, we all get that.
It is the subtext of every fangraphs article. We understand that Kershaw probably was never a true-talent 34% K pitcher, and that he’s probably not a tt 25% IFFB rate pitcher either. That said, this article still makes very valuable points, and backs it up with data that is actually meaningful, that being pitch selection and location data.
You’re right, everything is noise. FanGraphs is closed. Thank you, readers.
Does this mean FanGraphs closes and NotGraphs stays…?
I wish you could edit your comment to make it less assholish, too – then it would just be misguided instead of misguided and assholish.
What I like about the writer and his analysis is his subtle suggestion that pitching to miss bat and pitching to induce pop-ups is an excellent strategy. Since some fireballing pitchers find it hard to reach the 8th inning, it is fun to see how Kershaw is on a streak of 8+ IP starts where he has a better chance to control the outcome of a game rather than sitting in the dugout
Nice piece!
“but then strikeouts aren’t always automatic outs, themselves’
Any idea what the numbers are on that? What percentage of Ks result in the batter reaching on a dropped strike 3? I wonder if an IFB is truly as valuable as a K, or at least how close it is.
I guess the question is mainly just do hitters manage better than a .022/.022/.025 line on K’s?
I consulted the Play Index and used the Dodgers for an example. They have 1149 K’s this year and have had 4 players reach first on a WP/PB, 1 where I couldn’t tell what would have happened because first base was occupied, 1 where a runner scored while the catcher completed the out, and 31 other times the catcher completed the out at first with nothing else happening. So I think we can safely say that even with the possibility of a dropped third strike, a popup is a much better outcome for the hitter than a stirkeout, even if it is a terrible outcome for the at bat overall.
Also I just checked the Brewers for good measure, 3 safe at first(including 1 that was with first occupied but 2 outs), 1 with first occupied where I couldn’t tell, and 40 where the out was completed with no issues.
Who else hates him because theyve never owned him in a fantasy league EVER. Hes just so damn good. 99% of the males on the earth are jealous of him!