Clayton Kershaw’s Disappearing Fastball

Clayton Kershaw is nearing the end of another very good season. For the third straight year, the left-hander will fail to record 30 starts or 180 innings, but his 3.05 FIP and 2.51 ERA, his 25% strikeout rate and 4% walk rate make him one of the 10-15 best starters in the game by rate — and still a top-25 pitcher after accounting for volume. That’s not quite the 2011-15, Cy Young-level Kershaw who averaged more than 7.0 WAR per season, but it’s still good enough that he’ll likely opt out of his deal with the Dodgers in favor of entering free agency.

As to why Kershaw is now only “really good” instead of “Death Star-level dominant,” the easy culprits are age and health. He’s 30 years old now and has spent time on the disabled list due to back problems in each of the last three seasons. Perhaps directly related to those issues has been the loss of velocity on the lefty’s fastball. The graph below shows average velocity by season and includes his slider for reference.

For a decade, Kershaw sat at roughly 94 mph with his fastball. Last year, he averaged 93. This season, that figure is closer to 91. In the meantime, Kershaw has slowly modified his slider to increase its velocity into the 87- to 88-mph range we see today. (If you want to read more about the evolution of that pitch, re-visit Jeff Sullivan’s post on the matter from back in 2014.) The point here is that the slider, while perhaps experiencing a bit of a dip relative to last year, has exhibited pretty much the same velocity this season as the past few, while the fastball has slowed down significantly. The slider has been a pretty consistently very good pitch since 2014, with whiff rates in the mid-20% range and swings on half of pitches outside the zone. The whiff rate is down to 15% and swings outside the zone are closer to 40%, but the pitch is dropping a bit lower and inducing grounders on 66% of batted balls. Due to a high infield-fly rate, only 9.3% of batted balls are flies that leave the infield. The result for Kershaw on the slider has been a 47 wRC+ consistent with his career numbers.

The fastball has not remained as consistent. The last few years have seen batters swing and make contact on more pitches in the zone and take more pitches for balls. He’s given up 23 homers on roughly 2,000 fastballs the last two seasons after giving up about that many on more than 6,000 pitches in the previous four seasons combined. The fastball hasn’t been as effective, so Kershaw’s response has been the logical one: he’s throwing fewer fastballs. The graph below is a five-game rolling average since 2011 of Kershaw’s usage of both the slider and the fastball. For the first time ever, Kershaw’s slider is taking over.

This isn’t an unusual phenomenon, actually: pitchers are throwing more breaking pitches overall. Kershaw has been creeping in that direction for years, anyway, so the reduced velocity and effectiveness of the fastball have probably just hastened that trend. He’s still got the big curve, as well, and it is still getting him a lot of outs, with usage between 10% and 20% for his career, but the slider is the pitch taking place of the fastball. Here’s the same chart from above, except for this season only.

The first bars on the left include starts from the end of the 2017 campaign, and we see some movement back and forth in the first half of 2018. In the second half, though, a pretty clear trend emerges. Fortunately, for Kershaw, it doesn’t really matter how much he uses his slider — at least at these type of patterns — as there isn’t a relationship between slider usage and effectiveness, as the graph below shows.

Most of the time, the slider gets really good results, but there is no reason to think that bad outings have been the result of using the slider too much or that good outings have come from minimizing the use. Overall, the slider has been worth close to two runs for every 100 pitches. The fastball has been very close to zero. In theory, taking 10 fastballs every start and turning them into sliders like he has done over the past couple years is going to be worth about five runs over 25 starts or roughly a 0.30 difference in ERA. Kershaw could potentially move further away from the fastball, but the pitch can still be effective.

As Kershaw indicated on the game broadcast Saturday, he’s aware of changes he’s making, and he let’s hitters dictate which pitches are most effective.

John Smoltz: Clayton, for you and the adjustments you’ve made — I mean, just been an unbelievable career — and with age comes knowledge and with knowledge comes change. So how have you adapted to some of the changes you’ve had to do, not only this year but the last couple years?

Clayton Kershaw: Like you said, you just learn a little bit as you go. For me, I’m a little stubborn by nature and I’m only going to change if I have to. The hitters tell you when you have to change. When you go through a season, you really don’t feel like you are doing much differently, but then you go back and look and realize that maybe you are throwing certain pitches a little bit more. Maybe you are doing a little bit more or a little bit less on one side of the plate or another. The hitters will tell you what you need to do, and I’ve started to be more aware of that for sure.

Running his quotes above against fastball usage and effectiveness reveals an interesting trend, if we look at Kershaw’s starts since the beginning of May.

On the one hand, we might argue that, because his fastball is better when he throws it more, that Kershaw should be throwing it more (The relationship dating back to the beginning of the season isn’t quite as strong, perhaps due to randomness or perhaps due to pitching in a new world with a diminished fastball.) On the other hand, the more effective argument is that, when Kershaw’s fastball isn’t performing well, he limits its use and goes with the slider. We aren’t dealing with potential trends in usage just over the course of the season, but in individual performances. The better a fastball is doing in a game, the more he’s likely to use it.

There might be limits to how much Kershaw can go to his breaking pitches before he loses effectiveness, but he isn’t there… yet. Kershaw has made adjustments in the past, but most have been with an eye toward maximizing his across-the-board great stuff. These adjustments look geared more toward minimizing weakness. While a less effective fastball is never great for a pitcher, Kershaw’s ability to make these adjustments is an encouraging sign. So far, it appears to be working.