Johnny Cueto’s Changeup Problems

Johnny Cueto’s changeup is currently flatter than it’s been in five years. Literally. To some extent, everything in his arsenal is flat right now, but it’s most radical when you look at the changeup. With that wiggle in his delivery and a possibly falling arm slot, it’s easy to find a culprit. But he’s wiggled forever, why has he lost his drop now?

Take a look. If you ignore the one game in March last year, the last time he had a month with a changeup this flat was in August 2010.


Another thing that is going on right now is that Cueto’s arm slot flattened recently, and that could do something like kill your vertical drop on a pitch. Check out how he went haywire in September. The wiggle turned into a scratch on the record that month.


Check out the drop on all of his pitches in September compared to the rest of the year. Maybe everything flattened out as his arm dropped?

Johnny Cueto’s September Drops by Pitch
Pitch April-August Drop September Drop
FF 9.3 9.7
FT 6.7 7.0
FC 5.5 5.3
CH 1.8 2.9
SL 2.4 2.9
SOURCE: BrooksBaseball
Drop is in inches, and lower means more drop

Looks like everything is a little flatter, everything but the cutter had less drop this September. The changeup also firmed up a bit and his fastball velocity tailed off, so Cueto had his worst velocity gap on the pith that month. And yet… it was the fastballs that saw the biggest drop in whiff rates in September. His four-seamer, cutter, and slider saw their combined whiff rate drop by more than 50% that month.

In the past, I’ve found a link between release points and rise, so it’s easy to point to the arm slot as the source of the flattening. But even that research found a low r-squared (.21) that suggested that release point only predicted 21% of the variability in fastball drop. In other words, we may have found one-fifth of the problem here.

To see visually that we must be looking for other culprits, let’s look at Cueto’s changeup release point over time. While remembering that this has been Cueto’s worst month for changeup drop since 2010.


Throughout all of these ups and downs in his arm slot, the changeup dropped two to three more inches than it’s dropping right now. And yet, it’s still getting whiffs, and it’s the other pitches that have suffered more when it comes to whiffs. Cueto’s release point is famously variable, and has risen back to early-season levels, and yet October has seen the flattening continue: his sinker, change, four-seam, slider and cutter are seeing their worst month for drop right now.

But movement is best defined in a relative way. If your change and sinker had the exact same movement, it wouldn’t matter if they were ‘good’ in a textbook way, you’d still have two very similar pitches on your hand. And, since many of his pitches are flattening in the same way, Cueto’s difference in vertical movement is actually right in line with career norms. For the last three years, at least, he’s averaged four to five more inches of drop on the changeup than the sinker. September and October, he’s averaged 4.1 and 4.5 inches of additional drop on the change compared to the sinker.

If we go to the video, we’ll notice that Cueto’s change has actually added some movement with this new arm slot. Watch his September changeup (right) compared to an early-season change (left), both thrown for swings and misses. It was still a good pitch from the lower slot, just more horizontal than vertical.


Despite all of his shenanigans when it comes to his delivery, Cueto is actually very consistent when it comes to clustering his release points, even as his arm slot dropped. Last year, he stacked up well against Madison Bumgarner when it came to in-game and game-to-game consistency in his release points.

So, a guy with consistent release points but a crazy delivery and a falling arm slot has shown flatter movement on all his pitches late in the season — despite his arm slot recovering to early-season levels. Even as all of his release data has changed a bit, the relative movement between his pitches hasn’t changed much.

Perhaps we should look for an easier explanation. Perhaps we should take note that the two starts in which Cueto showed his worst fastball velocity were in late September and early October, and that you add in three more October and September starts if you look at his worst ten games by fastball velocity this year.

When we talk about seeing more life on Cueto’s pitches, we might be tempted to talk about drop. And yet, it might actually be more about the actual life on those pitches, as defined by the radar gun rather than by movement. Which counts as good news, then. Cueto sat at around 92.5 against Houston, which was the velocity he showed in July.

With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.

Newest Most Voted
Inline Feedbacks
View all comments
Concerned Reader John
6 years ago

It always takes me about 10 minutes to figure out the pitch movement charts. Because I’m dumb.

Concerned Reader John
6 years ago
Reply to  Eno Sarris

I’ve read explanations by you and other Fangraphs writers and they’ve been very helpful. It just seems to take me a minute – but maybe other readers have ideas. It’s probably one of those things where the best way to understand the numbers is to see them over and over again until they have context. That being said, I enjoyed the piece!

6 years ago

You’re not kidding. I think what threw me off was the “change-up hasn’t has drop this bad since August 2010,” when there appears to be a higher point in 2014. I get why you’d wanna throw out a single start with significant variance (the confidence interval is huge), but it makes it confusing.