Do sliders cause pitchers to lose curveball feel? MLB Pitchers weigh in
Originally published: Baseball Prospectus
GLENDALE, Ariz. — Shortly before the Nationals traded Lucas Giolito to the White Sox in a package for Adam Eaton, the 6-foot-6 right-hander had started working on a slider. He would debut the offering for the Charlotte Knights the following April after an offseason of refinement. To this day, it remains an important lateral compliment to the exceptional vertical break on his curveball. For a period of time while learning his slider, however, Giolito lost the feel for his curveball.
“If you’re so used to doing it one way to manipulate the spin on a ball,” Giolito said. “When you try and introduce the other [way], both of them can kind of get lost in the mix… you’ll end up throwing one pitch, which is kind of like a slurve--not very good.”
For Giolito, throwing a slider consists of staying on top of the ball and coming down the side to create spin and glove-side horizontal movement. For a curveball, he turns his hand over completely, like most pitchers, creating topspin rather than the side spin of a slider. The motions seem distinct, but muddling can still occur.
The loss of feel on a curveball after throwing sliders is not unique to Giolito. Sonny Gray has mentioned a similar issue when the Yankees acquired him from the Athletics in 2017. Prospect Luis Patiño battled the same issue when the Padres asked him to start throwing a slider upon signing from Colombia in 2016. Even Giolito’s teammate Dylan Covey, Padres opening day starter Eric Lauer and veteran Trevor Cahill confirmed their belief of the phenomenon.
Is the loss of feel for a curveball when increasing slider usage an odd coincidence, confirmed by a few players here and there? Or is there data to back up the loss of feel? And how important of a factor is velocity in the search for answers?
When Giolito starts his offseason throwing program, one of his focuses is finding the “feel” difference between his curveball and slider. He throws 20 light curveballs, 10 to 15 feet from his throwing partner, concentrating on the rotation of the ball off his fingers. Then he immediately switches to his slider, repeating the same exercise with a focus of being on top of the ball and engaging a “slight wrist action” at the end of the motion.
“[I] try to differentiate it as much I can to get that feel early,” Giolito said. “So when I start throwing hard and I start getting into a throwing program, [or] bullpens, I have that feel of what the difference is.”
Even at 24 years old, Giolito said he has enough experience to possess an innate feel for the difference between each pitch. If a quick lapse occurs, he knows where his mistake is and he can make a feel-based adjustment.
Trevor Cahill “spikes” his curveball by bracing the knuckle of his index finger on the ball to stay behind it better. He admits he has never had an issue with curveball feel, possibly because of this fundamental difference in grip. “I would assume it happens to guys who throw their slider and their curveball similar,” Cahill said. “With spiking it, it feels like a completely different pitch.” He admitted the loss of feel occurs when elevated cutter or slider usage creates a lapse with his sinker. His explanation echoes Giolito’s point about the blending of fine motor movements at the end of a complex total-body movement.
“I’m basically trying to lock my wrist out and throw [my cutter] like a four-seamer and [my wrist] just pronates so much on my sinker,” Cahill said. “My cutter is [difficult] because even when I let it go, I gotta like think about it for longer.”
Dylan Covey questioned whether pitchers who have a wider usage difference between their curveball and slider are more prone to loss of feel.
“I don’t throw my curveball enough to feel like I have an inconsistency,” Covey said. “It’s my last pitch that I use so it’s kind of always a little bit inconsistent just because I don’t go to it that much.”
In 2018, Covey threw his curveball 11 percent of the time, slightly more to right-handed hitters. Giolito, however, threw his curveball 10 percent of the time last season with a slight tendency towards left-handed hitters. Covey and Giolito have similar levels usage on their curveballs, yet the former thinks he does not throw the pitch enough to have an inconsistency, while the latter deliberately practices in the offseason to maintain his feel.
All three of these pitchers have thoughts that connect distinctly back to their repertoire and situation. Pitching is highly individualized and likely the reason why a direct answer to this question is so elusive. Pitchers have different grips, cues, mechanics, habits, and usage levels on pitches. Perhaps statistics can be the voice of reason to an individualized issue.
To determine a “change in feel” the BP stats department and I focused on two factors: change in movement and change in effectiveness. A change in movement can be defined as any positive or negative fluctuation of the horizontal or vertical break of a pitch. If a pitcher loses “feel” on a curveball, we would expect the pitch to change shape in some way.
A change in effectiveness is more difficult to determine. For purposes of this brief study, we focused on Fangraphs run values per 100 metric. Simply put, it helps us answer “how well does a pitcher perform using a certain pitch” (more info here). The issue with this measure of effectiveness for a pitch is the lack of an ability to provide anything more than a basic understanding of a pitch’s effectiveness over time. Looking at snapshots of an individual season can be extremely noisy, and often highly contextual: a pitch may be successful because it’s great, or because it isn’t, and so the hitters are sitting on some other pitch and get fooled. But let’s work with what we have and see if any of the information leads us to loosely spot a trend.
We considered all starting pitchers between 2008 and 2018 with a minimum of 100 innings pitched. We focused on pitchers who experienced a five percent increase in slider usage and a five percent decrease in curveball usage, an arbitrary point we felt could be interpreted as a change in approach rather than random variation. We ended up with 23 pitchers after some minor manipulation.
In terms of horizontal and vertical curveball movement between year one and year two, there is no discernible change. Taking the median of our sample’s movement provides us an average increase in horizontal movement of 0.2 inches and decrease in vertical movement of -0.1 inches--hardly an indicator of feel loss.
In 16 of the 23 cases above with a five percent slider increase and a five percent curveball decrease, the pitcher’s run value on their curveball fell in year two by an average of 1.7. Also understand that for 11 of these 23 pitchers, in year two, their curveball usage fell below eight percent, represented in the far right column above. Pitch usage below eight percent to either handedness makes it hard to consider a pitch more than an afterthought in a pitcher’s repertoire due to sparse usage. What we can discern from the run value changes is more thought provoking for specific cases rather than ground breaking. Take, for example, two pitchers from the sample above.
For Matthew Boyd, velocity is a key factor that adds context to this loss of feel concept. His fastball and curveball both fell around two mph between 2017 and 2018. He also drastically increased his slider usage by 21 percent but managed to keep his curveball usage stable due in part to a drastic cut in his changeup usage.
Boyd’s curveball gained more side-to-side movement and depth, but with the context of his velocity drop, it makes sense that a slower breaking ball will break both laterally and vertically more than a harder version. Also consider Boyd’s slider velocity is abnormally low, nearly four mph below the 2018 league average slider velocity of 84 mph. His slider is actually closer in velocity to the average curveball (78 mph in 2018) than the average slider.
Did Boyd gain feel on his curveball because it broke more? Or did he lose feel because the pitch, in that particular season, was less effective according to run values?
Noah Syndergaard’s case is the inverse of Boyd’s. Between 2015 and 2016, Syndergaard’s slider usage increased by 19 percent and his curveball took a 13 percent hit, but his whole repertoire increased in velocity. His fastball gained over one mph, his slider gained just over three mph and his curveball ticked up nearly two mph. With the velocity changes, he seemed to have reshaped his curveball as the pitch lost side-to-side movement and gained vertical depth.
Did Syndergaard lose feel on his curveball because the pitch’s run value fell slightly in the presence of increased slider usage? Or can the argument be made that a reshaping of a pitch in the presence of more velocity means the pitcher has an innate feel for the offering?
Velocity is a variable that makes determining the loss of feel particularly difficult, but it can also be part of the fix for rediscovering feel. “The only cue for me is to throw the slider a little bit harder,” Eric Lauer said. “Try to get it a little bit straighter.”
Lauer learned his slider during his freshman year of college at Kent State University in Ohio. He has been tinkering with grips ever since, on occasion using one variation for time intervals as short as one week. In instances where he loses feel for his curveball, he increases the velocity on his slider to reset himself. “They can have the exact same break for all I care,” Lauer said. “But as long as the velocity is like 10 mph of difference, then that’s what’s going to throw the hitter off.”
In 2015, Eno Sarris dug into sliders and discovered key characteristics by plotting horizontal and vertical movement as well as a pitcher’s velocity differential between his slider and his fastball. One of Sarris’ findings applies to Lauer’s personal cue: Velocity and drop are the most important variables of a slider, whether considering whiff rates or ground-ball rates.
Lauer’s conscious decision to throw his slider harder to regain feel creates a wider velocity difference between his curveball and slider. Conversely, it creates a smaller velocity difference between his slider and fastball. Based on Sarris’ study, the result of a smaller slider to fastball velocity difference would be more ground balls. Lauer may be manipulating his feel by gripping and ripping the pitch, per say, but the harder variation is actually a pitch with very different characteristics.
The cue for Lauer might be widening the velocity gap between his slider and curveball to make the pitches so distinct he could not possibly lose feel. But such a wide velocity difference could also create a noticeable arm speed difference, potentially leading to a tipped pitch. This cue is similar to a side thought Cahill had when talking about how he manipulates his slider.
“The separation and the velos [between my slider and curveball] for me are big enough where they don’t really mesh,” Cahill said.
Velocity matters and it might be the best way to regain feel, but unintended consequences may exist when too much tinkering occurs. This cue might be more applicable to the standard pitcher than Giolito’s offseason routine or Covey’s thought that repetition, above all else, reigns supreme.
Talking to pitchers and looking at data reminds us how unique pitchers are. It is difficult to take a simple phenomenon and obtain a sound conclusion without numerous qualifiers. According to some pitchers, the loss of feel on a pitch like a curveball can occur, but the fix is probably as unique as how the pitcher obtains their feel for a pitch. And according to statistics, the individualistic nature of pitching makes it hard to find a trend among a motley of arms with different pitch shapes, grips and velocities. Pitching is all about losing, except when it’s about finding, and ultimately, everyone has to find their own way.