Luis Castillo Has Revitalized His Season

I am not going to beat around the bush: Luis Castillo had a miserable start to the season. On Opening Day against the Cardinals, he allowed 10 runs (eight of which were earned) and recorded zero strikeouts in just three and a third innings of work. He bounced back about a week later against Pittsburgh, tossing seven innings with five strikeouts versus one walk without surrendering a run. But despite that showing, Castillo’s Opening Day struggles proved to be more than a blip on the radar for a pitcher who ranked among the top 20 in baseball over the three seasons prior to 2021. At the beginning of May, Castillo still had an ERA above six (6.07); it would not dip below that mark until his June 15 start against Milwaukee. His ERA peaked (apart from the stretch between his first and second start) at 7.61 on May 23. As of this writing, his ERA is 4.20; he has accumulated 3.0 WAR over 163 innings, placing him 26th in the majors among qualifiers.

So what has changed? Back on May 18, Justin Choi wrote about Castillo’s performance through his first eight starts, attempting to diagnose what plagued the Reds right-hander. Justin first looked at Castillo’s changeup, which had long been his most effective offering but which was generating far fewer whiffs than before. Justin found that the pitch was dropping more than it had in the past, though Castillo was still able to locate it just below the lower edge of the strike zone. The rest of Castillo’s repertoire (a sinker, four-seamer, and slider), on the other hand, was being placed right down the pipe. Justin concluded that the location of Castillo’s changeup were so different from that of the rest of his pitches that batters were either laying off the changeup or swinging at the few that landed closer to the heart of the zone.

Since Justin published his analysis, Castillo has compiled a 3.15 ERA and ranks 10th in WAR. Let’s try to locate the difference. Year-over-year, his pitch mix is largely the same:

Luis Castillo’s Pitch Mix
Season CH FF SI SL
2018 26.2 36.4 20.9 16.5
2019 32.3 29.2 21.5 17.0
2020 30.0 27.1 25.2 17.7
2021 29.2 29.0 24 17.9
SOURCE: Baseball Savant

Castillo has continued to lean more on his changeup and four-seamer against lefties (almost four out of every five of his offerings) and utilizes his sinker and slider more against right-handed batters (about two thirds of his pitches with the platoon advantage). But when you break 2021 up into two pieces, using the data from Justin’s article as the cutoff, you can see Castillo has made slight modifications to his mix as the season has progressed:

Luis Castillo’s Pitch Mix Since May 18
Time Frame CH FF SI SL
Pre-May 18 35.0 29.2 19.5 16.3
Post-May 18 27.3 28.9 25.4 18.4
SOURCE: Baseball Savant

The changes aren’t massive but they are worth noting. Castillo has been using his sinker more since his initial struggles, especially against right-handed opposition. His changeup usage is also down, possibly in response to the opposition not offering at the pitch a ton early on. These tweaks have extenuated the qualities that made Castillo’s pitch mix both diverse and deep.

Apart from his slider, the swinging strike rates on all of his pitches are still down from 2020. The changeup has decreased by 6.6 percentage points, the four-seamer by 3.3 points, and the sinker by 1.1 points. The slider is up to 19.5% from 16.2% in 2020 and is pacing his changeup for the first time in his career. Still, all those rates are above the league average, save for the sinker. Those swinging strike rates look better across the board when you split his season between before and after the third week in May:

Luis Castillo’s SwStr% Since May 18
Time Frame CH FF SI SL
Pre-May 18 15.9 9.0 3.8 10.8
Post-May 18 19.1 15.6 6.1 21.5
SOURCE: Baseball Savant

With regards to the location of his changeup compared to the rest of his pitches, Castillo has more aggressively attacked the zone with the former, forcing batters to offer at it more often and making it more difficult to differentiate from his fastballs and slider.

After pitching very aggressively with his fastballs in the early going, posting zone rates about three and five percentage points higher than the rest of the league with both pitches, Castillo has backed off a bit. His zone rate on the sinker has decreased to 55.4%, only two percentage points higher than the major league average. For the four-seamer, it is down to 52.1%, which is below his peers for that particular pitch type (54.4%). The slider is the pitch he has been primarily relying on for swings and misses out of the zone, only placing it across the plate 39.5% of the time since May 18, about 11% less than the league mean. Castillo needs to be able to locate his slider over the plate because, as a percentage of pitches out of the strike zone, Castillo’s slider is below the 50th percentile in chase rate. Thus, he cannot rely on the prospect of batters consistently expanding the zone when he deploys it.

Castillo’s willingness to throw the changeup in the zone (up about two percentage points when comparing before and after the end of May) has undoubtedly helped him induce more whiffs given the pitch’s bat-missing qualities. Justin noted in his piece that Castillo’s changeup had been dropping more than in the past and wondered if that, along with the pitch’s location relative to the rest of his arsenal, made it easier to pick up out of his hand. Castillo seems to have made some tweaks to remedy this predicament; his changeup is dropping 0.94 inches less since May 18 and has moved 0.9 inches more in the horizontal direction while adding 1.8 ticks in velocity. It now pairs better with his slider, which he is also throwing harder. The effect of that velo increase has been less movement on the pitch because there is less time for the Magnus force and any seam-shifted wake effects to take hold. The changeup and slider are now thrown within 2 mph of each other and have vertical movement profiles within an inch and a half but deviate by over two feet in terms of horizontal break. It comes as no surprise that the slider and the changeup are both getting Castillo swinging strikes on close to one out of every five pitches he’s thrown since he made these adjustments.

Justin also included a note he received from Michael Ajeto about some issues with Castillo’s release point consistency through his first handful of starts. On all of his pitches, he has released the ball closer to third base compared to the end of May, on the order of anywhere between an inch or two. The same goes for the height of his release; he is throwing the pitch lower to the ground now than he was before, though this is partially a result of adding some extension, which naturally makes you release the ball lower to the ground because you are driving down the mound more. With regards to the consistency concern, there has been very little difference in the spread of his release points in terms of each direction’s standard deviation (in inches):

Release Point Deviation by Pitch Type
Time Frame Pitch Type X Release StDev Y Release StDev Z Release StDev
Pre-May 18 CH 2.79 2.79 1.01
Post-May 18 CH 2.46 2.46 1.02
Pre-May 18 FF 3.44 3.44 1.20
Post-May 18 FF 2.78 2.78 1.30
Pre-May 18 SI 2.82 2.82 1.60
Post-May 18 SI 2.86 2.86 1.27
Pre-May 18 SL 2.98 2.98 1.12
Post-May 18 SL 2.75 2.75 1.29
SOURCE: Baseball Savant

Both the lack of change in Castillo’s release point and the consistency of those release points can be succinctly summarized in a single chart. You can see a lot of overlap in points before and after May 18; the spreads look almost identical:

Castillo has always been among the best starters at inducing groundballs. Since 2019, he has posted 56.6%, 58.4%, and 56.5% groundball rates (per Baseball Savant’s definition), with average launch angles allowed of 5.78, 2.20, and 3.63 degrees. He uses his sinker and four-seamer almost 60% of the time and both pitches get an above-average amount of horizontal break, while the sinker gets a ton of drop, making them both high-end groundball generating offerings. This season, Castillo initially struggled to get the number of groundballs he was accustomed to (though he still did so at an above-average clip), but a lot of his poor performance can be attributed to some misfortune in the batted-ball department:

Luis Castillo’s BBE Results
Player Time Frame GB% Avg LA wOBAcon xwOBAcon
Castillo Pre-May 18 52.6 5.1 .456 .376
Castillo Post-May 18 58.2 3 .354 .317
Rest of MLB Pre-May 18 44.5 11.6 .374 .372
Rest of MLB Post-May 18 43.7 12.1 .383 .365
SOURCE: Baseball Savant

Castillo’s wOBAcon allowed was more than 80 points higher than the league average despite the expected results being basically in line with those of his peers. When he did allow a fly ball early on, it seemed like it would always leave the yard; opponents posted an 18.9% HR/FB rate against him to start. That figure has regressed to a more manageable 13.9% since, just about the league average.

So, has Castillo made improvements as the season has gone on? I would say he certainly has. He fixed the issue of his changeup sticking out from the rest of his arsenal and has made more of an effort to throw it in the zone, which has bolstered its swinging strike rate along with the whiff rate of his slider. He has also induced more groundballs as the season has gone along, which is essential for a pitcher whose four-seamer does not have bat-missing characteristics and who throws a sinker about a quarter of the time. Castillo was also the victim of some batted-ball demons that have naturally corrected themselves as the season has progressed.

Both the natural regression and the small tweaks have enabled Castillo to perform like the pitcher he showed himself to be the past several seasons, one who is among the best in the league and a potential rotation anchor for a club with postseason aspirations. The Reds have been one such club the past couple of seasons, toeing the line between being in and out of the playoff picture for much of 2020 and ’21. With Castillo pitching the way he has, along with great seasons from Wade Miley, Sonny Gray, and Tyler Mahle, and a potent offense, the Reds will be a tough playoff out if they can make it into (and out of) the NL Wild Card game.





Carmen is a part-time contributor to FanGraphs. An engineer by education and trade, he spends too much of his free time thinking about baseball.

newest oldest most voted
Dmjn53
Member
Dmjn53

Will be interesting to see if he can ever harness the talent consistently. Feels like he’s still a kid but he’ll be 29 in 3 months