Hyun Jin Ryu’s Multiple FastballsBen Clemenson December 1, 2020 at 6:00 pm

Watch any amount of baseball these days, and you’ll see a familiar pattern: a catcher giving a pitcher a high target for his fastball. This is hardly a secret; it’s so obvious, in fact, that I don’t really know how to make a good introduction out of it. Did you know that pitchers throw four-seam fastballs high in the zone? You did! I don’t need to tell you that. But this article is about four-seam fastballs, so the paragraph feels necessary.

Anyway: Four-seamers work better when they’re thrown up in the zone. That’s not some silly contextual thing, or even really up for debate. It’s just how the pitch works. The backspin on the pitch means that hitters generally make contact under the center of the ball. Given that the normal launch angle for a ball up in the zone is already high, hitting under a high pitch means pop-ups and lazy fly balls. It also means whiffs.

Don’t believe me? I mean, first of all, just watch some baseball. The pitch that dots the top of the strike zone makes hitters look foolish with great frequency. You can also take a look at this table, which divides the strike zone into vertical thirds and looks at only four-seam fastballs:

Fastballs in the upper third of the strike zone are the best of both worlds. Batters miss them more frequently than the other two sections, and even when they make contact, they’re not doing so with authority. Fail to find the top of the zone, and things get worse. Fastballs that wind up in the middle third of the plate still miss bats at an acceptable clip, but if the hitter connects, you might be in for a bad time.

The bottom third isn’t clearly better or worse than the middle third in and of itself. Batters miss fewer fastballs down there; that’s bad! They don’t hit for quite so much power or overall value, though, which lessens the pain. Either middle- or bottom-third fastballs are the worst; which is worse isn’t immediately clear.

There’s one thing I didn’t mention, however, and it’s pretty obvious. The foot bone is connected to the ankle bone, and the middle of the strike zone is just below the top of the strike zone. Think of it this way: if you aim for the middle third of the strike zone, your expected result isn’t great. If you miss low, eh, no biggie; the two zones are pretty similar in expected results. Miss high? Amazing! You just backed into the best possible fastball location.

Aim low, on the other hand, and things aren’t great. If you hit your target, you’re throwing it in arguably the worst sector of the strike zone. If you miss high, that’s also bad; it’s more or less indistinguishable from where you were aiming. Miss low, and things get even worse; not because batters crush fastballs out of the zone low, but because batters swung at only 8.9% of four-seam fastballs that missed the zone low last year. Simply put, missing low is a nearly-automatic ball.

Why, then, would you aim low with a four-seam fastball? You might have a fastball that doesn’t behave like a prototypical four-seamer. Sinkers are at home down in the zone, and cutters can sometimes have success there. Perhaps your pitch is being mis-labeled. Short of that, though, it’s safe to assume that four-seamers that end up down there are misses.

Why aim for the middle? The top of the strike zone has one big drawback: miss high, and batters are likely to take. It’s not as bad as missing low — batters swing at roughly a quarter of four-seamers that miss high — but if you’re behind in the count, aiming too high carries a meaningful drawback. In other words, we can’t assume that every four-seamer in the middle third is a miss; it might just be a matter of a ball hurting more than the chance of some contact.

Again, it’s not quite as simple as saying that a pitcher who throws a four-seamer low in the zone just missed. Zack Greinke, for example, threw 21.1% of his four-seamers in the bottom third of the strike zone this year. Tony Gonsolin checked in at 20%. What I would guess, however, is that in the absence of knowing anything else about a pitcher, give me the one who avoids the bottom of the zone as much as possible with their heater.

The list of pitchers who wind up in that bottom third least often has some names that won’t surprise you. Josh Hader checks in at the 14th-lowest rate of bottom-third four-seamers among pitchers who threw at least 100 of them in 2020 (sorry for that word soup, but there’s really no other way to convey it). He located only 5.6% of his fastballs there. Nick Anderson was 13th-lowest at 5.3%.

Cubs reliever Jason Adam, who rode his fastball to a 36.2% strikeout rate, came in 10th at 5% even. He also displayed the downside of avoiding the bottom third: Pitchers who aim high will end up there with their mistakes least often, but aim too high or throw too wildly, and you might walk 13.8% of opposing batters, like he did.

In fact, this little gimmick statistic has almost no relationship with how good a pitcher was, and when I say almost no relationship, I mean it. It had a 0.002 r-squared with ERA-, 0.003 with FIP-, and 0.004 with xFIP-. In other words, basically none of the variation in those statistics could be explained by the percentage of four-seamers that those pitchers threw in the bottom part of the strike zone. Looks like my guess from up above was wrong!

Why do I care, then? Partially because it’s December 1 and you’re reading a daily baseball blog; you’re clearly interested in weird things like this. I’m not judging — I’m writing it, after all — I’m just saying that minor things can still be worth articles. There’s another reason, though: A few names atop the list don’t fit the mold of fireballing relief types, and those names are quite interesting to me.

Sixth on the list is Hyun Jin Ryu. The Toronto ace has been aiming higher in the zone of late:

His average fastball is four inches higher now than it was from 2016 to ’18, which is a meaningful jump. He’s also been masterful those past two years. Coincidence? Well, basically yes. Ryu was great in 2018, too, so it’s not like something he did between ‘18 and ‘19 was the difference. One thing that did change, though: Starting in ’18, Ryu added a sinker to his repertoire, and starting in ’19, he started to use it more while simultaneously creating more vertical separation between the two fastballs.

By creating this separation and commanding both pitches, Ryu found a way to get the best of both worlds with his fastballs. His sinker gets more grounders, and his four-seamer gets more whiffs, so he used them that way: He threw most of his sinkers in the first two pitches of an at-bat, when batters are less likely to protect the plate by swinging at anything in the zone. He also threw a few in 3–0 and 3–1 counts as get-me-over pitches. He might want to work on that next year — only five of the 11 he threw there actually hit the strike zone — but that was pretty clearly the plan.

When whiffs were necessary, it was four-seamer time. Given that he frequently turned to the pitch when he had a ball to spare and was looking for a whiff, the top of the zone seems like the best place to throw it, and as I mentioned above, that’s exactly what he did. In two-strike counts, he threw 92 four-seamers and eight sinkers, and as we already know, he threw that four-seamer higher on average than he ever had before.

It’s not obvious whether Ryu’s plan had a beneficial effect on his four-seam whiff rate, or really any beneficial effect at all. Oh, you could certainly find some beneficial effects. For example, take this little zinger: In plate appearances where he threw a four-seamer but not a sinker, Ryu allowed a .407 wOBA in 2020, and in plate appearances where he threw a sinker but not a four-seamer, he allowed a .441 wOBA. When he threw both, he allowed a .349 wOBA, far better than either alone. Shazam!

Only, in plate appearances where he threw neither, he allowed a .213 wOBA. And of course, plate appearances with more pitches have more chances for him to throw both, particularly plate appearances where he falls behind in the count and starts throwing fastballs to get back into it. On the other hand, a first-pitch fastball that the opponent put in play is necessarily only one of the two, and those are some of the most valuable plate appearances for batters. Confounding variables are everywhere!

There’s this, though: The three highest whiff-per-swing rates on four-seam fastballs in Ryu’s career have come in 2018, ’19, and ’20. His two highest GB/FB ratios have been in ’19 and ’20, largely due to the sinker. Each of the pitches has impacted his arsenal in a positive way, which is pretty much the whole point.

Should other pitchers try to duplicate Ryu’s multi-fastball ways? I’m skeptical. Not just anyone can start throwing a sinker without losing the feel for their primary fastball. Heck, Ryu’s primary fastball might in fact be his cutter, which he throws more often than either of the other two, though personally, I see that one as more of a slider given its movement and velocity. The point is, though, that “make every pitcher do what Ryu does” simply isn’t a workable plan.

Are there any takeaways from this article? The main one, for me, is that if you can manage it, giving pitches specific roles can work. It’s more complicated than that: How each fastball interacts with your other pitches matters, and you shouldn’t add a pitch just to add one, but done with a plan, using your whiff-inducing pitch to induce whiffs and your grounder-inducing pitch when you wouldn’t mind a grounder can flatter both of them.

That junk statistic, by the way? It ended up not mattering at all. Ryu’s genius isn’t that he happens to avoid throwing fastballs in the bottom third of the zone. It’s merely a byproduct of something else cool he’s doing, and that’s a worthy finding in my book.
Read More