Changing Breaking Balls, By Movement

Yesterday, we looked at the biggest changes in fastball movement (rise and sink) in starters and relievers from their offerings in 2014 vs. this season. Today, we’re going to do the same thing with breaking balls (pitches known for their movement), which should hopefully yield some interesting takeaways as we move forward with this young season.

As I said in the previous article, more movement doesn’t always mean better results: it can be a catalyst for some changes in peripheral numbers, however, and can point toward raw improvement in a pitch. We’ll go into some information related to whiff rates and batted ball profiles with these breaking pitches, looking for any change in production that goes along with change in movement.

The standard preface: all stats are farmed from Baseball Prospectus’ PITCHf/x leaderboards. It’s obviously still very early, so take these results with a grain of salt, and mostly as something interesting to watch as the season progresses. Today we’ll divide these pitches by curveball and slider, looking at starters and relievers together. We’ll also divide sliders by lefties and righties, as the movement data is obviously quite different for each of them. As a baseline, I’ve used a 50 pitch minimum for both starters and relievers in 2014.

Rick Porcello is coming up a lot in these increased movement datasets: he ranked 3rd yesterday in increased rise on his four-seamer, and here he is again, leading all pitchers in increased drop on curveballs. That extra movement has seen his whiffs/swing increase 17.2% over last year, even though hitters are not swinging as often at the curve this year (-12.9 swing rate). It is also interesting to note is that he’s taken about a mile per hour off the curve while maintaining his fastball velocity so far this year, trying to stretch out the difference between the two pitches. The bottom line: Rick Porcello is trying different things this year, and it remains to be seen how it will turn out.

Let’s take a look at an average curveball from Porcello this year:

A lot of the other pitchers on the list are curveball specialists, punctuated by Clayton Kershaw: he’s always been one of the leaders in vertical movement on his curve, but he’s taken it to the next level in 2015, adding an inch and a half of drop. His -10.2 inches of movement compared to a pitch with no spin is only bettered by two other pitchers in the majors this year, Chris Tillman (-10.25) and Mike Fiers (-11.95). Kershaw has seen his whiffs/swing on his curve jump almost 14%, up to to an obscene 51.1%. League average is around 28%.

Last week, I wrote about how Sonny Gray has gradually become a very different pitcher since he was first called up in 2013, altering his pitch usage and movement over time. Part of his new approach is having two breaking pitches (or a big enough difference in the way he throws his curveball to register as two distinct pitches), with one of them classified by PITCHf/x as a slider. That pitch has really increased in horizontal movement this year, becoming a serious weapon in the process: a 21.8% increase in swing rate (he ranks first in the majors this year in that category) and a 13% increase in whiffs/swing puts him firmly in the top tier of quality sliders.

Here’s what that looks like, from the end of April:

After Gray, we have a diverse mix of front-line starters and relievers, the most interesting of whom is probably Drew Storen. Like Gray, he’s throwing a flatter slider this year with less vertical and more horizontal break — a possible driver of a 7% uptick in swing rate and a 27% increase in whiffs/swing. Those peripherals are the main reason hitters are 1-14 in at-bats ending in his slider with nine strikeouts so far this year. However, Neftali Feliz is a good reminder that performance is a lot more than increased movement, as he has taken away all of the vertical break in his slider for more horizontal movement. Whether it’s the main component or not, he has 20.6% less whiffs/swing to show for it. His other pitches aren’t getting missed much either, though.

This is a good time to refresh ourselves with the mechanisms of PITCHf/x, as negative movement values mean the pitch moves toward a right-handed hitter, and positive values move away from them. Since we’re talking about sliders thrown by left-handers, we want to look at the highest negative value changes.

That being said, Aaron Loup has decided to start throwing his slider very differently so far this year, adding almost three miles per hour and lots of horizontal break. The results so far: his swing rate is way up (+13%), but his whiffs are way down (-13.5%). It’s tough to know how any pitch is going to shake out long term given the current sample sizes, but by the numbers, Loup’s change has made his slider a less effective strikeout pitch that gets hammered when thrown poorly. He gave up one extra base hit on his slider all of last year, and he’s already given up two in 2015, one of which left the yard. Sure, that could be luck: it’s a tiny sample. Most of the time, however, tinkering with a pitch that already works during the season doesn’t bode well for pitchers.

Here’s Loup’s new slider:

Interesting to note: once Chris Sale’s go-to offering, his slider has all but fallen into total disuse, and one would have to think it might be part of a plan to try to keep him healthy. Sale is only throwing his slider 9.2% of the time so far this season, down from 18.4% last year and the 29.6% rate he had in 2013. It’s also been a very poor pitch when he has thrown it, with the usage patterns indicating he basically now uses it only when ahead against left-handed hitters. He’s becoming a fastball/changeup guy, which makes sense given his great velocity and everyone believing in his injury risk. Andrew Miller is also a name that stands out on this list, as he’s posting a whiff rate 8.6% higher than last year on his slider. Considering the quality and success of Miller’s slider in 2014, it’s not a surprise he’s throwing it more often, at almost half of his pitches so far this season.

We can gain insight into possible changes that pitchers are making to their offerings with this data. Much of it may be noise, yes, but there is signal here, and it tells us that pitchers tinker, change grips, and try to gain an edge by honing the subtle movement of their pitches. From the look of it, sometimes those experiments work. Most often, the ones that don’t are gone before we can even notice.