Every year, around this time, MLB Network does their Top 10 Right Now list for each position, including starting pitchers. This year, Luis Severino wasn’t on any of the lists I saw, which would have been inexcusable just six months ago, as he was dominating for the entire first half of the season. However, in the second half of the season, Severino was a complete shell of himself — a nightmare every time he took the mound. So it’s really no wonder why he hasn’t been included on any of these lists. He needs to earn his way back on.
Through his first 18 starts (March 29-July 1), Severino was one of the most electric pitchers in baseball. He posted a 1.98 ERA, with 138 punch outs in 118 innings of work. He allowed a slash line of .195/.251/.282. In his last 14 starts (July 7-September 25), Severino had a 5.65 ERA, struck out 82 hitters in 73 innings, and hitters slashed .299/.340/.515 against him.
Naturally, I had to figure out how this total fall from grace happened and if there is anything Severino can do to make sure this doesn’t carry over into 2019 — because the Yankees cannot afford for that to happen.
On the surface, it was tough to figure out what exactly happened to Severino in his last 14 starts of his 2018 campaign. He wasn’t surrendering more walks, his strikeout rate didn’t fluctuate too much, he didn’t lose any velocity, and his swing-and-miss rate was nearly three percentage points above league average. It certainly was not an obvious problem.
Digging a little deeper into the numbers, I found something odd.
Severino’s fastball got absolutely hammered in the second half of the season. Hammered.
Opponents hit his fastballs 35 feet farther, on average, and nearly 2.5 mph harder. You can also see from this graph that hitters were swinging and missing on his fastball less and less each month — it even dipped as low as 13 percent in July.
But why? Why all of a sudden were opponents raking whenever Severino threw his four-seamer?
Nothing had really changed about his fastball — he was throwing it with the same velocity, it had the same spin rate, and he was using it the same amount as he was in the first half of the season when hitters were missing it and finding little success against it.
Essentially, after gathering all this data, I had a couple theories and wanted to test them out to see if they held any weight.
Theory 1: Severino lost his slider (his other plus pitch), which allowed hitters to sit on his fastball.
To test Theory 1, I wanted to see if there was any significant change in movement on his slider between the first and second half of the season. I also wanted to see if hitters had more success against his slider, too, in the second half of the season. Finally, I wanted to see if hitters were chasing his slider as much in the second half of the season as they were in the first half.
In Severino’s first 18 starts, his slider was dropping nearly 36 inches (with gravity) and breaking well above 6 inches horizontally. In the second half of the year, he had lost 2–3 inches of vertical drop and roughly 1.5 inches of horizontal break. In baseball, that’s a pretty substantial difference. It just wasn’t the same pitch as it was in the first half of the season. For what it’s worth, his average spin rate on his slider in the first half of the year was 2,911 RPM (the best rate in the league during that time), compared to 2,818 RPM over his last 14 starts.
Losing a few inches of break, vertically and horizontally, on a slider can completely throw a pitcher off. It means it’s not crossing the plate where he is expecting it to cross, leading to more missed spots.
In Severino’s first 18 starts, his slider was lethal; opposing hitters hit just .152 against it, with a .181 wOBA and a .217 SLG. During this time span, that .217 SLG on a slider was good for fifth among starters, and the .152 batting average was the eighth-best mark in the league.
In his last 14 starts, hitters saw much more success against his breaking ball, hitting .257 with a .306 wOBA and a .431 SLG. The .257 average was the highest in the league (tied with Jameson Taillon) against a slider during this time frame, and the .431 SLG was the second-highest mark among all starting pitchers.
The loss of break on his slider in the second half of the season — essentially making it an entirely different pitch — lead to more success for opposing hitters. In addition, hitters weren’t missing his slider as much outside the zone in the second half of the season.
Inevitably, he started throwing it out of the zone more often and started throwing the pitch less frequently each month, as he continually lost confidence in it.
With only three true pitches in his repertoire (four-seamer, slider, changeup), throwing the slider less means he needs to get people out with his fastball. But when you throw your slider less, and when you do throw it, it’s not as good as it used to be, that means hitters can sit on the fastball and drive it all over the field.
I’m not entirely sure how to fix his slider. It could be something as simple as a tiny adjustment in his arm slot or his release point. Perhaps he started slightly twisting his wrist at the point of release. Maybe he should play around with the grip, but I would hesitate messing with a grip that once produced such a lethal slider before looking into the other options. Whatever the fix may be, I’m confident that his slider is not permanently broken, nor is it gone for good; this is definitely an issue that can be fixed.
Theory 2: Severino fell behind in the count too often, which allowed hitters to sit on his fastball.
To test Theory 2, after seeing Theory 1 certainly had a lot of validity, I wanted to see how often Severino was falling behind in the count, and also how early he was doing so. First pitch strikes are certainly important, but the three first pitches together are the most important. Starting 2–1 versus starting 1–2 makes a world of difference and can greatly tilt the odds in the pitcher’s or hitter’s favor depending on which way the count leans. Ideally, then, pitchers should be able to throw (all pitches) for a combined strike rate of 66+ percent.
So, let’s get into the numbers.
In the first half of the season, Severino had a first pitch strike percentage of 70.5 percent — that’s really good. In the second half of the season, Severino threw a first pitch strike 66.1 percent of the time. That’s still good, but also still a four-point drop-off.
I went a step further to see how Severino responded after throwing a first pitch ball — did he even the count back up at 1–1 most of the time and shift the power back on his side or did he let the count slip to 2–0 and hand the hitter a huge advantage? The answer is yes and yes.
In the first half of the season, after starting 1–0, Severino’s next pitch was a strike 74 percent of the time — four percentage points higher than his first-pitch strike percentage during that time. In the second half of the season, after starting 1–0, Severino’s next pitch was a strike just 60.3 percent of the time. That’s a pretty big drop off and not the best number you could have for that statistic. When down 1–0, you should be able to come in with a strike to even the count back up 70–75 percent of the time.
Over Severino’s last 14 starts, when the count sat at 1–0, 2–0, or 2–1, Severino came in with a fastball 54.7 percent of the time. In those counts, opponents hit .615 against his four-seamer (they hit .478 against his fastball in the same counts during the first half of the season, likely because they were still weary of the slider). When the count was 0–1 or 1–2, opponents hit a modest .270 against his fastball, yet they hit .344 against his slider (as opposed to .163 and .169 against the fastball and the slider, respectively, in the first half of the season).
After diving into and testing both theories, it’s apparent that both of them hold some validity. However, it appears to me that Theory 1 had a trickle down-effect of sorts and is partly what caused some of the issue I proposed Severino had in Theory 2.
Losing his slider caused Severino to put more emphasis on his fastball, which he threw for strikes less frequently the longer the season went on and the longer he went without having faith in his slider. This, in turn, allowed hitters to tee off on his fastball when he got behind in the count. Losing his slider also hurt him when he was ahead in the count, because he wasn’t getting hitters to chase it as much, so they could lay off of it and wait for him to throw a fastball.
The secret for whether or not Severino’s poor second-half performance carries over into this year and he has a disastrous 2019 is hidden in one thing: his slider.