Clayton Kershaw: Visualizing Greatness

As if you needed more proof of Clayton Kershaw's greatness. (via Arturo Pardavila III)

As if you needed more proof of Clayton Kershaw’s greatness. (via Arturo Pardavila III)

Greatness in sports is exemplified by athletes who reach the pinnacle of their sport, yet somehow improve upon their performance and reach a new level. It is a testament not only to their tremendous natural talent and passion for the game, but also to their intense competitive drive.

For eight consecutive seasons, LeBron James improved his field goal percentage, peaking at .567 in his final season with Miami, a season after winning back-to-back championships. Stephen Curry, fresh off his first NBA championship and being the driving force behind a fundamental shift in the way basketball is being played, is attempting three more three-point shots per game, all while shooting a career high from the three-point line. Following four neck surgeries, Peyton Manning could have retired as a first ballot Hall of Famer, but chose to return and had the most prolific season in NFL history.

This is not a story about basketball or football. This is about an individual who comports himself in a manner becoming of someone in the public eye. This is about a major league pitcher who has had the honor of having his starts called by Vin Scully every fifth day, who entered the majors as a 20-year-old with only a fastball that was not yet dominant and a curveball that had so much movement, umpires had trouble calling it for strikes. Clayton Kershaw’s brilliance, narrated beautifully by the inimitable Scully, has been very enjoyable over the last seven years; Watching without Vin Scully in 2017 won’t be the same.

“Let’s Get Back to This One”

In 2013, Kershaw posted his first sub 2.00 ERA – the original fielding semi-independent pitching metric. Naysayers would have been quick to point out that his 1.83 ERA was not fully supported by a 2.39 FIP and a 2.88 xFIP, despite Kershaw having outperformed his ERA estimators in every season since 2009 to that point. In 2014 he improved upon a 1.83 ERA by posting a 1.77, all while striking out more batters, walking fewer batters and inducing more ground balls (and to perturb the analytically inclined, I shall point out that he posted a vastly improved 21-3 W-L record).

Not satisfied with back to back sub 2.00 ERA seasons, Kershaw improved on his already elite skill, boosting his swinging strike percentage and posting the lowest zone contact percentage of his career. What is at the root of this other-worldly ability to improve on a season which only a handful of pitchers could hope to achieve? Let’s dig a little deeper.

We begin with a visual that helps illustrate the three stages that Kershaw’s career trajectory has undergone.

I’ve preselected Kershaw to highlight his seasons in the bottom chart; feel free to click around and view other pitchers below. This is a simple chart which graphs ERA to strikeout/walk ratio, with the size of the circle representing the pitcher’s strikeouts per nine innings, giving a visual emphasis to pitchers with equivalent K/BB but more strikeouts. Keep in mind, the data include only seasons since 1988, when Kershaw was born; during that time span John Tudor was extremely successful, despite a K/9 of roughly four.

What I love about visuals like these are the interesting outliers that pop out, the otherwise unknown and obscure name that jumps out at you for an interesting statistical quirk. What you’ll notice with Mr. Kershaw (excluding his rookie season, which didn’t hit the threshold) is that he’s had three distinct levels of play: 2009-10 with a K/BB of 2 to 2.5, 2011-13 with 3.5 to 4.5 and 2014-15 with 7 to 8. When you mouse over the bubbles close to Kershaw’s 2014/15 seasons, you’ll notice he’s surrounded by Greg Maddux, Roy Halladay and Pedro Martinez.

These graphs default to Kershaw vs righties (you can flip the filter to show lefties, or both), illustrating the growth in his game over the years. Plenty of lefties are tremendous against same-handed batters, but with Kershaw this dominance extends to righties as well. I want to point out here that my SwStr% numbers differ from the FanGraphs numbers (this may be due to my inclusion of foul tips as they track the FanGraphs numbers, but are consistently a few ticks higher).

From 2009 to 2015 his slider has gone from 12.7 percent SwStr% all the way to 28.7 percent, his curveball from 8.1 to 21 percent. His fastball declined a little in the early years, but has ticked up significantly in 2014/2015, likely due to improvements in his curveball and slider. I want to note that while this mirrors the general trend of SwStr% increasing year to year, the macro trend is closer to 0.2 to 0.4 percent a year, which Kershaw is far outpacing. What we see is a clear trend of a great pitcher improving his pitches year in and year out.

Kershaw may have hit his ceiling when it comes to inducing ground balls with his pitches, which may be reduced due to the increases in whiffs (a strikeout is necessarily not a ground ball). However, we can clearly see the trend from 2009 to 2013, when he continually improved his ground ball percentage; the continued improvement in the key aspects of his game are what I find so fascinating. We are clearly witnessing a pitcher who is constantly improving and polishing his game. With respect to quality of contact, his pitches are all pretty equal, suggesting that batters are not able to pick up on any of his pitches, unable to “tee-off” on any pitch, a sign of a pitcher with three dominant pitches.

If you flip the chart to lefties, you’ll notice that Kershaw’s fastball drew about 20 percent swinging strikes in 2015, which is just plainly ridiculous. There is a lot of noise in the data, but it does appear that his fastball has become more hittable as of late against left-handed batters, perhaps something to work on in 2016…

So how does Kershaw pitch to lefties and righties?

A Hardball Times Update
Goodbye for now.

The chart above shows where in the strike zone the median pitch from Kershaw will end up, with negative horizontal being outside for lefties and inside to righties. What is amazing is the tremendous consistency with the fastball; in every season after 2009, he’s kept his median fastball location to lefties within a range of 2.5 to three inches — essentially the width of the baseball. Kershaw clearly has a plan with the fastball to lefties — throw it up and away. You’ll notice that his median fastball is not thrown “down.” Rather, they are consistently above the mid-point of the strike zone, which is better for swinging strikes.

He throws his sliders down and away, with the median slider essentially living at the bottom outside edge of the zone, where it is nearly impossible to do much damage. In 2009/2010, he threw his curveball toward the outside edge of the plate, but has since brought it more toward the center of the plate, presumably to induce more swings and get more called strikes. It’s important to note that with pitches with tremendous vertical tilt, it may be beneficial to cross the plate slightly higher so as to give the visual impression of crossing the heart of the plate while still above the hitter’s knees.

Let’s look at Kershaw vs. righties:

I want you to look at the sliders, represented by triangles, and note the three distinct clusters. The 2009 slider is roughly belt high, the 2010-2013 sliders cluster 0.4 feet lower and interestingly, the 2014/2015 sliders are a further 0.4 feet lower in the zone. What is readily apparent is not only the year-to-year consistency, but that he is executing a very consistent game plan year to year.

The fastball tells the same story as the ones to left-handed batters: extreme consistency year to year and a preference for throwing up in the zone, with hitters unable to do much against them. He’s clearly tweaked his curveball strategy over time, going for a thigh-high curveball in 2009-2010, followed by a buried curveball in 2011, to what is now probably the optimal median location for a curveball, in the sweet spot where you induce ground balls, get whiffs and maximize called strikes.

If the best pitcher in the game has a distinct strategy of keeping the fastball up, why do teams consistently preach “keep the ball down”? Just asking…

I hope you’ve made it this far in the article, since I’ve left the best for last — an interactive dashboard where you can see pitch location and result, by count and pitch type, for every pitch Kershaw has thrown since 2009.

There are an assortment of ways you can slice and dice Kershaw’s pitches and I encourage you to play around and leave some insights you discover in the comments section. Note that these data include only Kershaw’s three main pitches and specifically exclude change-ups and pitches that PITCHf/x couldn’t classify. I’ve defaulted the visual to showing 3-1 counts (click on the 3-1 count to clear the filter) where Kershaw has given up an astonishingly low number of four home runs. Let me put that into context:

In 420 pitches thrown in a 3-1 count, Kershaw has thrown 403 fastballs. Of those 403 pitches only 11! have gone for extra base hits. 11!

In fact, in three-ball counts, he has thrown 1,345 fastballs and given up only 28 extra-base hits. You’re almost guaranteed a fastball in 3-0 and 3-1 counts, but are unlikely to do much damage when it’s coming from Kershaw. This to me is the true signature of greatness — even in extreme hitter counts (2-0, 3-1) Kershaw is still nearly unhittable, despite throwing 95 percent fastballs.

I’m looking forward to what 2016 Kershaw will bring to the table. If his trend of constant improvement continues, the best may be yet to come.

Eli Ben-Porat is a Senior Manager of Reporting & Analytics for Rogers Communications. The views and opinions expressed herein are his own. He builds data visualizations in Tableau, and builds baseball data in Rust. Follow him on Twitter @EliBenPorat, however you may be subjected to (polite) Canadian politics.
Newest Most Voted
Inline Feedbacks
View all comments
6 years ago

I’m a Giants fan, but I think Kershaw is the best I’ve seen — and I saw a lot of Koufax, including the Marichal/Roseboro game.

6 years ago

In baseball tribal knowledge…you pitch left handed hitters up and right handers down. Most left handed batters have a natural uppercut and right handers swing from top down.

6 years ago
Reply to  brian

Wouldn’t it make much more sense just to pitch each individual hitter according to how he swings?

6 years ago

Easier to digest: Top Lifetime ERA’s of pitchers born in the last 100 years (100+ GS):

Kershaw: 2.42
Ford: 2.75
Koufax: 2.76
J. Palmer 2.86
Seaver 2.86
Messersmith 2.86
Marishall 2.89

Although his ERA could still rise as he gets into his mid-30’s. The 33/34 point gap from the formerly best ever is not an accident. This was noted on MLB during the Dodger broadcast April 15. He has started 245 games–Koufax started 313 lifetime and Ford 438.

6 years ago
Reply to  Morgan

In this case, where you are comparing across such different eras, it’s important to adjust ERA. The mid-60’s and the late-90’s are like night and day when it comes to run scoring. ERA+ is scaled to 100, and increased by how much better (or worse) that pitcher’s ERA is compared to average, as a percentage. By that measure, the top two:

Kershaw: 155 (meaning 55% better than league average for his ballpark and era)
Pedro: 154.

Kershaw is slightly ahead, but considering he hasn’t had his decline yet, it will be quite unlikely that he maintains it, unless his peak is longer than we think it will be, allowing him to increase his lead in the meantime. This is how the rest of your list pans out, with ERA+ and career rank:

Ford: 133, T-26th
Koufax: 131, T-35th
Palmer: 125, T-64th
Seaver: 127, T-47th
Messersmith: 121, T-100th
Marichal: 123, T-83rd

The thing about a lot of these guys is that, even if they don’t have the greatest adjusted ERAs of all time, the fact that they maintained numbers significantly above average over such long careers is exceptional. Kershaw has established such a great baseline that he could be merely average for another 10 years and still show up on the list with these greats.

6 years ago

Is there any chance of seeing those K/BB charts but with K-BB? I think it is pretty well established that the latter is more meaningful than the former, especially around guys like Kershaw who are at the absolute top end of the K% ranks. K/BB is also susceptible to being tricked by extreme low-BB guys. Phil Hughes was good in 2014, but not all-time good.

Joe W
6 years ago

Nice article!
You wrote, “What is amazing is the tremendous consistency with the fastball; in every season after 2009, he’s kept his median fastball location to lefties within a range of 2.5 to three inches…” Could we get some context for that to understand the consistency? What is the range of average median fastballs that other pitchers have thrown to lefties/righties over the past several seasons? I assume the range would be larger, but for all I know, other pitchers in the league are just as consistent in fastball location, albeit consistently more hittable than Kershaw.

Ben Johnson
6 years ago

When I saw that low number of home runs off Kershaw’s curveball, I was kind of curious to see if my favorite was included.

It wasn’t, because most of the scripts people use to harvest Pitch f/x data don’t pull postseason unless you specifically ask for those dates individually (at least that’s true for the pitchRx package in R).

And the brooksbaseball data:

6 years ago

In my mind, considering the degree to which pitchers have improved since the 1950s and 60s, Kershaw is the best pitcher of all-time. Not the greatest, but the best. By that I mean that you put him side by side with any pitcher ever and he’s the better one in terms of “stuff” like fastball velocity and movement, location, deceptiveness, competitiveness, sequencing, and off-speed offerings.

Possibly the greatest three year run ever until he got hurt this year.

Thanks for the article,