Further Investigation of Justin Steele’s “Fastball”

by
June 27, 2023

Last year, Justin Steele verged upon pitch-tracking-era history:

If you don’t know how this ends, my exceptionally dim-witted-but-nevertheless-talented colleague and friend Alex Fast jinxed it in spectacular fashion:

Nevertheless, the streak remains interesting because Steele is up to his same antics.

Only four pitches have been thrown 500 times this year and allowed just one home run. Here they are, in order of total pitches thrown and accompanied by total plate appearances (PA) completed:

  1. Steele’s four-seamer (725 thrown, 199 PA)
  2. Kevin Gausman’s splitter (622, 160)
  3. Jordan Montgomery’s sinker (584, 170)
  4. Hunter Greene’s slider (516, 100)

It’s one thing to accomplish this feat at all; it’s another all together to do so with a fastball (kudos to J-Mont, but a sinker ain’t a four-seamer). Here is the top of the list of only fastballs that have allowed one or fewer home runs, ordered by most thrown. The gulf between first and second from a volume standpoint is astounding:

  1. Steele’s four-seamer (725 thrown, 1 HR)
  2. Shohei Ohtani’s four-seamer (427, 1)

And, to put a cherry on top, here’s all pitches thrown at least 1,500 times since the beginning of the last year, ordered ascending by home runs allowed. Steele’s fastball has allowed half as many home runs as the next-best pitch:

  1. Steele’s four-seamer (1,880 thrown; 3 HR)
  2. Gausman’s splitter (1,591; 6)
  3. Alex Cobb’s sinker (1,547; 6)
  4. Logan Webb’s change-up (1,502; 6)

Despite allowing only two home runs last year, Steele’s fastball got hit around, allowing a .377 batting average on balls in play (BABIP). Things have changed this year, though, its performance improving dramatically. Among 48 pitches that have incurred at least 100 batted ball events (BBEs) this year, Steele’s fastball ranks 2nd-best in exit velocity (EV) suppression (-3.7 mph), behind only Drew Smyly’s curve (-4.6 mph). Next-best to Steele is Corbin Burnes‘ cutter at -1.9 mph. It’s a Grand Canyon-sized margin, and it becomes all the more spectacular when you remove breaking pitches from the equation.

To heap further praise, it ranks 12th in deserved ERA (dERA) and 16th in predictive FIP (pFIP) among 75 pitches thrown at least 500 times. It can be reckless to make determinations of pitch quality based on ERA estimators, but I find them to be instructive. And, based on these estimators, the pitch grades out well from both descriptive (backward-looking) and predictive (forward-looking) perspectives.

It would stand to reason, then, that such a high-performing pitch would also grade out strongly by pitch models or, at least, through pitch values. Looking first at pitch models, PitchingBot grades Steele’s fastball a 63 comprehensively (stuff and location) on the 20-80 scouting scale, which is, by definition, a “plus” pitch. That’s good! However, Pitching+ (Stuff+ and Location+ combined) gives it a 98 (where 100 is average), so the jury remains out on its quality.

Pivoting to pitch values, Statcast ranks Steele’s fastball 10th, with 9.2 runs saved. That’s really solid. But that’s, like, half the runs saved by Joe Ryan’s four-seamer, which, to me, in appealing to my own emotions and not my logic, feels wrong. Pitch values err on the side of description, and Ryan’s .200/.240/.300 slash line against his four-seamer includes a gaudy 25.4% strikeout-minus-walk rate (K-BB%). Ryan doesn’t really change the usage of his fastball in two-strike counts, whereas Steele throws his considerably less in those situations. That kind of difference in usage will adversely affect an offering’s total pitch value simply by virtue of volume.

Perhaps more interestingly, though, Pitch Info doesn’t even rank Steele’s four-seamer. That’s because Pitch Info doesn’t think his four-seamer is a four-seamer at all—it says it’s a cutter. I didn’t believe it at first, primarily because Baseball Savant calls it a four-seamer. (It’s a known bug, although some might call it a feature, that the MLBAM team allows pitchers to label their own pitches, almost regardless of the actual movement profiles, which creates taxonomy issues.) My Pitch Comps tool validates Pitch Info, comping Steele’s four-seamer to several cutters. Returning to Pitch Info’s pitch values, Steele’s now-it’s-a-cutter rates very highly. But I don’t care about that part anymore.

Before I continue, though, it’s imperative that I acknowledge that I’m not even the first at FanGraphs, let alone elsewhere, to highlight Steele and his unique arsenal—that distinction belongs to our Jake Mailhot, who incidentally used my Pitch Comps tool to yield a couple of cutter comps way back last summer, too. (He even owned Fast, and who can resist a chance to own that guy?) Bryan Smith of Bleacher Nation beat everyone to the punch when he wrote almost exactly one year ago about Steele’s fastball’s evolution, and Brett Taylor, also of Bleacher Nation, further highlighted the pitch this January. There very well may be others I’ve missed, too. I’m late to the party, obviously, perhaps not even fashionably so.

But I’ll be damned if I can’t show up late to a party and not find a way to enjoy myself. Because I love pitches that do this, that blur the line between two or more pitch classifications. Such blurriness signifies a pitch that defies typical classification, yes, but also probably defies convention, defies expectation. And a pitch, through its velocity and release point and movement profile, that defies convention with an atypical shape can find unconventional success. Such pitches deserve additional celebration! Let me celebrate!

To do this, though, I need to recalibrate my calculations. Among 61 cutters thrown at least 100 times (as of June 22), Steele’s cut-four-seamer-whatever would rank 4th in average velocity (92.4 mph) and 6th-lowest in vertical approach angle (VAA) above average (-0.89° “VAA(AA)“). But for the Pitch Leaderboard I calculate VAA(AA) on my Pitch Leaderboard relative to the pitch type–that is, relative to other four-seamers. If I recalculate Steele’s VAA(AA) relative to cutters, it becomes 6th-highest (+0.76°). Its velocity betrays the approach angles you’d expect: it should be a flatter, period. But, instead, it’s too steep to be a four-seamer, its lack of induced vertical break (IVB) creating the kind of drop you’d expecft from a sinker, and it’s nearly too steep to be a cutter. It straddles several classifications that can’t quite pin it down. Is it a four-seam cut-sinker? I don’t know. I don’t know.

I say this a lot, and I’ll say it again here: sometimes it’s better to be different. It’s one thing to chase the qualities of a Good PitchTM—we have pitch models and pitching labs to thank for that. It’s another thing, though, to simply be an outlier. (Ironically, pitch models may frown upon this.) Most pitchers’ repertoires feature extreme amounts of cut or carry (“ride,” whatever you want to call it) but not both. Steele indeed achieves variations of both, creating extreme cut and drop that contribute to a FIP-beating profile few others currently replicate from a specs standpoint.

If we could add a third dimension to this plot, one that measures velocity, Steele’s whatever-this-is would find itself even more isolated.

If the visual repositories in hitters’ brains are stored with tens (hundreds?) of thousands of pitches they’ve seen in their lifetimes, it makes sense that a pitch whose shape declines to resemble any of those pitches will confound opposing hitters. And that’s exactly what Steele’s cut-four-seamer has done this year, continuing a trend of elite home run suppression and finally tapping into the EV-dampening abilities that seemed achievable (as evidenced by said home run suppression) but eluded him last year by weaponizing absurdly low IVB relative to its velocity band and Steele’s higher arm slot.

(I’m not certain of this explains why the pitch succeeds other than claiming its uniqueness is the X-factor. Sorry! That’s all you’re getting from me.)

There’s not enough time in the world to devote this amount of attention to every pitch in every pitcher’s arsenal. It’s just not possible, especially when you are the parent of an almost-three-year-old. For fantasty baseball purposes, I’m obligated to make some mental shortcuts when evaluating players, if for no other reason but to preserve my sanity, for better or worse. Apparently a faulty, or at least vulnerable, heuristic of mine is dismissing pitchers who throw their fastballs too often, and especially those who appear to have gotten too lucky with those fastballs (for Steele, that luck refers to his home run suppression). Most fastballs are bad, and non-aces who throw too many fastballs often find themselves in trouble.

Steele’s fastball isn’t bad, though—it’s quite good. And evidently it may not be a fastball at all, at least not traditionally speaking. Whether or not it’s a fastball—and it really threatens to not be a fastball—it features a unique combination of physical specs. And if a pitch boasts unique specs, it may find itself defeating hitters in ways other pitches can’t.

And Steele’s four-seam cut-sinker-thing does exactly that.





 

Two-time FSWA award winner, including 2018 Baseball Writer of the Year, and 8-time award finalist. Featured in Lindy's magazine (2018, 2019), Rotowire magazine (2021), and Baseball Prospectus (2022, 2023). Biased toward a nicely rolled baseball pant.

Comments are closed.