by Matt Wilkes

Luis Castillo and his mission to develop another put-away pitch

Spring (or, in the case of 2020, summer) is a time for pitchers to not only get in line with their mechanics but also to work on improving pitches or trying new ones out. For Luis Castillo, the slider was the area of focus. He threw it heavily in his spring outings in an effort to develop a second put-away pitch along with his devastating changeup. We read the same story before the 2019 season.

At its best, Castillo’s slider can make hitters look like this:

The slider gives Castillo a pitch that moves differently than any of his three others. His four-seamer, two-seamer, and changeup all tail toward his arm side — that is, toward righties and away from lefties. The slider does the opposite, giving him an extra level of deception.

The GIF below is a perfect example of this in action. It’s not a particularly beautiful slider from an aesthetic standpoint. It doesn’t break much at all, actually. But Pete Alonso likely expected the ball to tail back over the plate — after all, Castillo’s other three pitches do. The result: Alonso swings through a pitch nowhere near the strike zone and looks rather foolish in the process.

Cherry-picking a couple of highlights is fun, but how did Castillo do in his mission to improve his slider in 2019? Where does can he still get better? And why is it such an important pitch for him? Let’s break it all down.

The improvements Castillo has already made

Castillo dropped his xwOBA with the slider from .301 in 2018 to .291 in 2019. Its vertical movement improved from 0.7 inches above average to 1.8 inches above average. The slider’s whiff rate (47.5%) rivaled his changeup’s (48.0%), and Castillo raised its chase rate from a poor 29.9% to an average 34.2%. It held hitters to an average exit velocity of 83.5 mph, lowest among all his pitches. He threw it at roughly the same rate as he did in 2018 (16.9%), with most of its usage coming against right-handed hitters (24.8%).

In a development that should surprise no one, Castillo increased the slider’s spin by over 100 rpm — from 2,206 to 2,330 — in his first season pitching for Derek Johnson. While its spin efficiency was only 10.4% — fifth-worst in baseball — that isn’t necessarily a bad thing for a slider. Only 72 of 506 pitchers who threw at least 250 sliders in 2019 had a spin efficiency above 50%. Low efficiency means Castillo’s slider gets a lot of “gyrospin” — which does not lead to movement — instead of “sidespin” that leads to sharp horizontal break (picture Trevor Bauer’s slider for a good example of high spin efficiency and lots of sidespin). When visualizing gyrospin, think of the way a football or bullet moves.

This goes a long way toward explaining why Castillo’s slider moves the way it does. The slider gets above-average vertical movement (1.8 inches above average) or “drop.” Gravity obviously plays a large part in this. Unlike a four-seam fastball, a slider doesn’t usually have much backspin, meaning the Magnus Force has little effect on it and it drops much faster — hence why it can be effective when paired with the heater. However, Castillo’s gyrospin slider gets almost no horizontal movement. It gets 6.8 fewer inches, or 107% less break, than average — breaking only 0.4 inches on average.

The following GIF from Rob Friedman shows this close up — watch how the ball essentially spins in a tight circle and drops straight down:

So, why is Castillo’s slider effective despite limited horizontal movement? In most cases, a pitcher’s fastball is more important than how much his slider breaks. Pitchers can succeed with a slider that has high or low spin efficiency if it has deception. The more a slider looks like a fastball, the more likely it is to fool a hitter. It may be more apropos to say the longer a slider looks like a fastball, the better.

Castillo does this beautifully with his two-seamer and changeup, a concept called “tunneling.” It’s a lethal combination, especially against left-handed hitters, because the ball darts away from them and the pitches look almost identical. He tends to throw the two-seamer and changeup in the same parts of the zone, down and away from lefties, making it even harder for the batter to discern what’s coming. With a nearly 10 mph difference between the two pitches, the result is often ugly for the hitter.

The heat maps demonstrate Castillo’s efforts to disguise the pitches as one another:

Castillo often throws his four-seamer further up in the zone and on the opposite side of the plate from his typical two-seam and changeup location. This makes the slider, which he predominantly likes to throw down and away from right-handers, a natural tunneling partner for the four-seamer.

As the increase in chase rate may indicate, the gyrospin on Castillo’s slider probably helps him. Since Castillo’s slider doesn’t break horizontally, it may have the appearance of a four-seamer even longer before it drops below the zone. The horizontal movement on Castillo’s four-seamer — which is 3.8 inches above average — can make the pair even more deceptive.

Eric Jagers, Reds assistant pitching coordinator, showed this in a video in 2018:

Castillo generally likes to throw the four-seamer and slider in the same location, which helps his tunneling efforts:

That doesn’t necessarily mean he tunnels the slider with his four-seamer exclusively, however. It can play off his two-seamer and changeup as well. When he does that, it’s usually against lefties. The next GIF, also courtesy of Rob Friedman, demonstrates this perfectly.

All three pitches start low and in the middle of the zone. The slider drops below the zone and slightly toward the batter. The sinker and changeup — which, in another example of tunneling, look almost identical — break away from the batter.

Here’s a recent example from way back in spring training:

Where Castillo’s slider can get better

Yet, the breaking ball has some flaws. Hitters managed a .291 xwOBA against it in 2019. That certainly isn’t bad when compared to, for instance, the average fastball. But it lags behind the league-average slider (.269 xwOBA), suggesting further room for improvement.

Although his slider didn’t result in much hard contact or allow hits often (.190 batting average, .209 xBA), it did get punished when he left it up and out over the plate (.452 slugging percentage, .426 xSLG, .262 isolated power, .217 xISO). For a reference point, the league average xISO for sliders was .147 and .177 for all pitches. He allowed six home runs on the slider, one more than his changeup — which he threw twice as often. This is potentially where a lack of horizontal movement hurts Castillo. If his slider doesn’t get its typical drop, it’s essentially a straight-as-an-arrow, 85-mph, batting-practice pitch if it’s left over the plate.

Five of the six home runs he allowed with the slider came in the last two months of the season. While this may seem like random variance, evidence points to something being off, whether it be mechanically or with his pitch grip. As the season wore on, Castillo got less and less drop on his slider (note that the upward trend in this graph equates to less movement):

Over the last three months of the season, Castillo’s slider lost about an inch of drop compared to the first three months of the year. This may not seem like much, but it can make the difference between a whiff and a barreled ball.

This change was also paired with better control. Castillo threw his slider in the strike zone 34.8% of the time from March through June and 42.4% thereafter. That’s a potential plus if he wants to keep hitters honest! Even with the decreased movement, it equated to more swings (50.6% the last three months vs. 35.6% in the first three months) and whiffs (49.6% vs. 44.7%). But the slider was also hit harder as he threw it in the zone more, as lapses in command came back to bite him. In the last three months of the season, his slider was hit for an 85.7 mph exit velocity and .347 xwOBA; in the first three, the slider held hitters to an 82.4 mph exit velocity and .232 xwOBA.

Castillo is also not all that comfortable throwing the slider against left-handed hitters. Of his 519 sliders thrown in 2019, only 156 were against lefties or 9.7% of his pitches. That said, he did make some strides with his slider when he did throw it against lefties, increasing its whiff rate from 32.1% in 2019 to 53.3% and lowering its xwOBA from .344 to .293.

Why the slider is important for Castillo

At the most basic level, the slider gives Castillo four dependable pitches in his arsenal and, thus, more variables hitters need to account for. It also gives him a pitch with a completely different movement profile than his four-seamer, two-seamer, and changeup. Those three pitches all move to Castillo’s arm side — toward right-handed batters and away from lefties. The slider, as previously mentioned, has almost no horizontal movement but above-average drop.

The chart below demonstrates the stark difference in horizontal movement between his four pitches.

Most importantly, though, it gives him a second putaway pitch along with the changeup. Castillo began heavily featuring his changeup last year, throwing it more than any other pitch. That makes a lot of sense — it is his best pitch, after all. But, as pitching coach Derek Johnson said, hitters began to sit on the changeup as the year went on.

A devastating swing-and-miss pitch, the changeup became more hittable late in the year. Granted, hitters still didn’t do much with it when they put it into play, but his whiff rate on the pitch dipped.

Make no mistake about it: a whiff rate above 40% is still elite. But it’s difficult to get whiffs half the time, as he was doing early in the season, when you throw a pitch 32% of the time and you’re not a reliever. More balls in play — even if they’re not hit hard — means more chances for bad luck to come into play, whether it’s a ball finding a hole or a fielder making an error.

Castillo’s slider has shown an ability to register an elite whiff rate of its own. Refining it and throwing it a bit more can keep hitters from sitting on his changeup, especially in two-strike counts. In other words, Castillo’s repertoire can go from dangerous to lethal.

[Photo: https://twitter.com/Reds/status/1143552512089370624]

Matt Wilkes got hooked on Reds baseball after attending his first game in Cinergy Field at 6 years old, and he hasn’t looked back. As a kid, he was often found imitating his favorite players — Ken Griffey Jr., Adam Dunn, Sean Casey, and Austin Kearns — in the backyard. When he finally went inside, he was leading the Reds to 162-0 seasons in MVP Baseball 2005 or keeping stats for whatever game was on TV. He started writing about baseball in 2014 and has become fascinated by analytics and all the new data in the game. Matt is also a graduate of The Ohio State University and currently lives in Columbus. Follow him on Twitter at @_MattWilkes.

4 Comments

  • Brian Van Hook

    Thanks for this analysis, Matt! … Side question: Does Derek Johnson call pitches from the dugout, or leave it to the catchers? I ask because it would seem like the catchers platooning would mean both need to know how often to go to the slider when they might be tempted to go back to the established changeup.

    • Steve Mancuso

      The catchers call the game. They look over when runners are on base to see if there is a pick-off call. The problem isn’t with Barnhart or Casali knowing the pitchers. That’s their full-time job and what they focus on. But the third catcher – Farmer right now – might not be as up on it. But they talk before games and every start with Castillo. That’s also the coach’s jobs to make sure the catchers are prepped and know what the pitcher wants to and should be throwing.

  • Anonymous

    This is awesome analysis and wildly intriguing. Keep up the great work. Hard to find this deep level of analysis elsewhere.

Leave a Reply