Trouble with the curve

This post is also available in / Disponible en: : enEnglishesEspañol (Spanish)

Sandy Koufax, Bert Blyleven, Satchel Page, Tom Glavine, Barry Zito, Clayton Kershaw: all of them amazing pitchers with a big common thing: they own some of the best curveballs in the game.

We are used to seeing Kershaw’s 12-to-6 beautiful curveball, like this one, in action:

That was Nick Ahmed being Kershaw’s 2500 lifetime strikeout a little more than a month ago using, of course, a great curveball.

Or this one, handling the upper zone without the fear that should come when doing it:

By the way, this is the exact moment when Christian Yelich realized he was dead:

Yep, dead.

So… Why doesn’t Clayton Kershaw use his curveball more?

According to Baseball Savant, Kershaw’s pitch usage since 2015 during regular season games looks like this:

His average use of the curveball since 2015 is 16.6% with a wOBA of just .171 on that pitch for the period, ranking 3rd among pitchers with at least 1000 curveballs thrown during that span. This regular season he had an 18.8% of curveball use, which is the highest in his career, batters hitting a .228 wOBA against it.

So, again, why doesn’t he uses it more? Well, as usual in baseball, there is not a simple evident answer to this.

Pattern logic leads to the belief that if you do something successfully, it means you should repeatedly do that to be successful all the time; the problem with that logic in pitching is that you have to account for batters ability to adapt to those patterns: if all they see are curves, they will surely adjust pretty fast and those nice hanging balls will probably end in the bleachers.

That’s why pitchers know (or should know) they can’t use curveballs in every count or in all counts. To illustrate this let’s look at the wOBA leaders sin 2015 on any 2 strikes counts while throwing a curveball (min. 5000 pitches, 200 CB):

Those are pretty nice numbers; Kershaw is a solid 12th with 0.146. The only pitchers that meet these qualifiers with a wOBA north of .300 are Julio Teheran and Clay Buchholz with .318 and .309 respectively.

Now, let’s check the results for the same qualifiers but in any favorable count for the pitcher (ahead):

Among the 109 qualifying pitchers in this situation, only 4 had a wOBA of more than .300. Again, Kershaw is in a great position with .153.

We can get the idea, if you have a good curveball and you are ahead or in a 2 strikes count, it will probably be good to use it.

Now, what happens when the pitchers got behind in the count? Let’s check it out:

That’s an abysmal difference; remember, in the previous charts only a handful of pitchers allowed a wOBA higher than .300; this time, only a handful of players allowed less than .300 and from there, wOBA simply skyrockets.

Now let’s be clear, the qualifiers are skewing the results because the chart is showing pitchers with a minimum of 200 CB thrown in that specific situation (pitcher behind in the count) so that includes pitchers that even not being successful continued throwing curveballs, which is one hell of a recipe for disaster.

In that regard, Kershaw knows better: during that whole span he only threw 28 CB when falling behind in the count, batter’s wOBA was 0.501 in those opportunities.

That is the difference between good and extraordinaire: you do what you’re good at in the right moment; that’s one of the reasons Clayton Kershaw is indeed extraordinaire.

All data used was taken from, and/or, unless otherwise stated differently. pCRA data was taken from this Tableau, maintained by its creator Connor Kurcon.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.