The More It Stays the Same, the Less It Changeups

Last time we saw Tobias Myers, he was beating the bejeezus out of the New York Mets in Game 3 of the NL Wild Card Series: Five innings, 66 pitches, five strikeouts, no walks, two hits allowed, one hit batter. Being a rookie starting pitcher in a playoff game in the 2020s, he exited early, but having allowed zero runs, it was to a chorus of Hosannas mixed with “Who the heck is this guy?”

Myers’ chief weapon in his most famous start was his slider, which he threw 31.8% of the time, generating four whiffs, two foul balls, and three called strikes. The Mets made contact seven times, resulting in six outs.

But Myers’ most effective pitch during the regular season — at least on a per-pitch basis — was his changeup. Myers threw 245 of those across his 138 regular-season innings, resulting in a whiff rate of 44.4% and an opponent batting average of .083. Only four of the 27 changeups that were put in play turned into hits, and because all four of those were singles, Myers’ had an opponent SLG to match his opponent batting average: .083.

In the 2024 regular season, 1,249 distinct pitcher-pitch type combinations got used at least 200 times. Of those, Myers’ changeup had the second-lowest opponent SLG and the third-lowest xSLG.

Both Stuff+ and PitchingBot see Myers’ changeup as at least average, but in neither case is it really spectacular. PitchingBot has it 36th among pitchers who threw 100 or more innings last year. Stuff+ has it 17th, but with bottom-of-the-barrel location and Pitching+ grades.

That’s understandable, because just in terms of movement, Myers’ changeup looks pretty dire out of context.

If you go to the pitch movement section of Myers’ Baseball Savant page, you’ll see something that looks like the flag of Liechtenstein turned upside-down: a thick red stripe over a thick blue stripe. Let’s just look at Myers’ changeup against his two primary fastballs. You’ll find that Myers’ fastballs have some of the best vertical movement in the game, while his changeup is almost literally the opposite:

Except, like the flag of Luxembourg, elements of this table need to be reversed. (I worried briefly that I’d lost my handle on that simile, but I think we can all agree that everything is under control now.)

“Induced vertical break vs. average” is a great concept, but while the axis under consideration remains the same, which direction we’re talking about changes from pitch to pitch. Fastballs have “rise” (in an idiomatic, rather than literal sense — I know the physics, don’t @ me), so a positive IVB number means less relative drop. Changeups, like most other pitches, are designed to sink, so a positive IVB number means more relative drop. That -7.6 inches is relative to induced break for your average right-handed changeup; what it actually represents is nearly a foot of spin-induced rise.

So does Myers’ changeup break less because it’s bad, or does it break less because it’s supposed to be a rise pitch to match his fastball?

Based on where he throws it, the latter seems to be the case. Myers threw 22.9% of his changeups in the top four inches of the strike zone or higher in 2024; out of 130 pitchers who threw 200 or more changeups, only Brent Suter went soft upstairs more frequently.

This is in contrast to Myers’ slider usage. Myers usually buries his primary breaker, which is entirely normal. A high four-seamer with rise, followed by a breaking ball in the dirt — that’s downright traditional pitch usage.

I was curious whether the other elite changeup artists in the game also play off their fastballs in this fashion. So I took the 15 pitchers who threw at least 200 changeups in the 2024 regular season, and posted either an opponent wOBA or an opponent xwOBA (or both) of .200 or lower. It’s an eclectic group: lefties and righties; starters and relievers; All-Stars and… um, non-All-Stars alike. Hard throwers, soft tossers, fat ones, skinny ones, short ones, tall ones, one fish, two fish, red fish, blue fish:

So you see, about half of these guys throw multiple fastballs. Sánchez is sinker-only, but Myers is well within the norm in that he throws a four-seamer and a cutter and very occasionally a sinker. But is it normal for a changeup specialist to have a hard-rising changeup like Myers’? After all, isn’t the whole point of a changeup to, well, change?

Sure, but direction isn’t the only thing that can change. Most of the pitchers on this list throw a changeup with an observed movement direction that goes pretty hard to the arm-side. One way to put it is this: Baseball Savant expresses this with a clock. Everyone but Myers has a changeup that moves in a direction that would represent an inappropriate time to eat lunch: 10 o’clock or earlier for lefties, 2 o’clock or later for righties.

Except Myers, whose changeup comes in at 1:30 and overlaps with his four-seamer.

In short, Myers already has that high four-seamer/low slider combination that’s been the bread and butter of relievers since time immemorial. Hitters can only cover one of the two. If they guess fastball, they might be on the money in terms of movement, but with such similar movement on the changeup, they could swing only to discover that the pitch hasn’t arrived yet.

That must be frustrating.

And that does seem to be what’s happening. Here are the relationships between fastball and changeup in terms of IVB and velocity. I’ve marked Myers in yellow, and the two pitchers whose most common fastballs are not four-seamers — Sánchez and Kelly — in red and purple, respectively:

Myers’ approach is not unique. (Alex Vesia, for one, does the same thing with his fastball and changeup, though he throws the latter pitch even less frequently and effectively than Myers does.) But it is unusual.