Robert Gsellman blasted out of the gate this season, performing like an elite reliever for the first month of the season as the Mets jumped out to a strong start. As the team has faded, however, so has Gsellman, who now finds himself with an ERA and FIP much closer to last year’s underwhelming campaign. In a lost season, this prompts a major question: Is Gsellman a late-inning arm that can potentially inherit the closer role from Jeurys Familia after this season, or is he another middling right-handed reliever better served as a taxi squad arm?
To start, here are Gsellman’s monthly splits this year.
So, what’s going on here? After a disappointing 2017 in which Gsellman pitched through injuries as a starter, he transformed into an elite reliever for 15 innings, then went right back to being bad. Let’s assume for the moment that he’s not just hurt again and try to pick out what’s changed beneath the surface.
Those splits are a little all over the place. First, Gsellman’s strikeouts went away in May. Then they mostly came back in June, but batters started teeing off, hitting more fly balls, and increasing their hard-hit rate. July was a mix of both problems, with the strikeouts disappearing again and the fly ball rate staying high, albeit with a more reasonable hard hit rate.
What was Gsellman’s early season plan? Arm-side sinkers:
Glove-side breaking balls:
Basically no four-seamers or changeups. This was a really effective attack for Gsellman, who induced a ton of whiffs down in the zone:
As well as a plethora of ground balls:
Something must have changed at the start of May. It wasn’t Gsellman’s velocity, pitch movement, or release point—his velocity actually improved—but his zone profile did change a little. Ignoring his changeup for the moment, here’s a comparison of his zone profile before and after May 1:
It’s not a glaring shift, but Gsellman definitely threw more pitches over the plate as well as up in the zone, and that’s a bad strategy when you’re relying on a sinker/slider/curve mix. This shift in location wasn’t the only flaw, as batters started swinging and missing a lot less down in the zone:
So to recap, Gsellman threw his best pitches in worse locations, while the ones he kept in the optimal portions of the strike zone were less effective. The ‘why’ behind these changes isn’t immediately clear. Perhaps after being overworked in the first month of the season—15 innings, which is not egregious but high for a recently-converted starter—Gsellman’s control slipped and he’s never recovered. Perhaps he developed bad habits that tipped his pitches in some manner. Maybe he’s pitching hurt again, as he did last season. This is all speculation.
Unfortunately, the most likely explanation is that Gsellman’s early season success is simply a product of small sample sizes and that his performance over the three months since is more reflective of his true talent. We’ve seen Gsellman throw some incredibly nasty sinkers and sliders that leave hitters totally baffled, and he was ranked as the 17th-best prospect in baseball by Baseball Prospectus coming into the 2017 season. However, the movement on his sinker isn’t remarkable. Ten inches of arm-side run is good, but not elite, while the 5.54” of vertical movement is middle of the pack. The velocity isn’t elite, either, and his slider is one of the straightest in baseball.
For the moment, let’s push forward and assume that Gsellman can regain some of his early season success by tweaking the location of his three primary weapons, and move on to the changeup. He tried to work the pitch in more after April, and it hasn’t worked at all. Batters have slugged .692 against Gsellman’s change, and it’s been literally one of the worst pitches in baseball in terms of Fangraphs’ pitch value (-3.44 wCH/C). This was a pitch he succeeded without early in the season, and while experimentation is always worthwhile, particularly for a young player in a lost season, admitting that said experiment is not working is also important.
The final observation here is not something that Gsellman’s done but rather something he hasn’t. In 2017, he used his bad four-seam fastball roughly 19% of the time, and batters slugged .651 against it. He’s used it only three percent of the time this season, and batters are still slugging .625 against it. Take a look at where he’s thrown it:
It’s primarily been a weapon to attack left-handed hitters inside, though he grooved it far too often for the pitch to be effective. Here’s the whiff rates the pitch generated, and you can probably figure out where this is going:
Unsurprisingly, Gsellman’s four-seam fastball is most effective up in the zone. This is a point that’s been brought up repeatedly in recent years as hitters gear their swings more and more towards fly balls and home runs: guys are vulnerable upstairs. Gsellman generates below-average rise and doesn’t have elite velocity, but a straight fastball up in the zone is an interesting foil to the breaking balls and sinkers down in the zone. Pumping its usage up into the 10% range and focusing on using it almost exclusively up in the zone would be a worthwhile experiment for the remainder of the season.
So, in conclusion, here are three things for Robert Gsellman to tweak over the remainder of the season:
- Work on keeping the sinker and slider in the lower third of the zone as much as possible.
- Stop throwing changeups. It’s a bad pitch, and it’s really not a necessary weapon in a relief role.
- Experiment with four-seam fastballs up in the zone.
At this point, the jury is still out on Gsellman, but I worry that his by-the-numbers unremarkable stuff gives him a relatively low ceiling. That doesn’t mean he can’t be more effective then he’s been since the start of May, and I believe the three changes above will help him improve. Hopefully he mixes things up and proves he can a late inning fixture over the last half of the season.