Not every pitcher needs a changeup

Updated Mar. 4, 2020 11:39 p.m. ET

share

facebookxredditlink

foxsports

The pro-changeup argument is unassailable. Still, there are many many pitchers out there who can't manage the pitch. And not every single one of them is destined for the bullpen. There is an alternative. 

Pitchers who throw the changeup a lot might have the best injury outcomes. Based on research done by Jeff Zimmerman, below is a table that shows the disabled list percentages for starters based on their favorite pitch. Looks like there's something healthy about the changeup. For each bucket, we tried to use cutoffs that led to similar samples, so the slider-heavy pitchers (starters who use the slider more than 30% of the time) throw more sliders than the changeup pitchers (starters who use the change more than 20% of the time) throw changes, but that's just because there are fewer heavy-change pitchers. 

Injury rate by pitch

Beyond health, there are plenty of reasons to promote the changeup as many organizations do. They bust platoon splits by offering a pitch that breaks in a different direction than sliders and curves, at least. And they go slower than sliders and faster than curves, so they also offer a change of pace. 

But there is one pitch that can do many of these things almost as well as the straight change. The roundhouse curve. 

Only two pitches have significant reverse platoon splits: the big curve and the straight change. The curve is the slowest pitch, too. A pitcher who had a slider (or cutter) alongside a 12-to-6 curveball would theoretically have a platoon-busting arsenal that featured three different speeds. That would leave arm health as the only concern, and there is no real magical solve for that issue. 

Are there pitchers out there who should ditch the change and focus on their slider and curve? It worked so well for Garrett Richards that there have to be others out there who could follow his lead. 

In an effort to judge changeups, there are two ways to go about things. One method, outlined by Harry Pavlidis, is to look at the shape of the pitch and its velocity difference from the fastball. His findings, in short, were that movement is great either way, but the velocity gap is more important if you use the pitch for whiffs than if you use it for grounders. We can reference these findings as we take a look at some key pitchers below. 

Let's use a slightly different method to find the pitchers who should ditch their changeups, though. Let's try to give individual pitches a score, and then use that score to find the worst changeups in baseball. 

By focusing on the "cleanest" outcomes -- groundballs and whiffs -- we can remove some of the noise that comes with looking at ball-in-play results. Instead of asking someone to judge what a line drive is, we're asking if the ball went along the ground. Instead of trying to judge which hits were actually hits and not errors, we're just asking if the batter missed the pitch. 

After correlating the changeup's swinging strike and groundball rate to the pitchers' overall ERA, we can find a way to weight grounders and whiffs. Whiffs are more important than grounders just as strikeouts are more important than groundballs -- the strikeout is an out 99% of the time and the grounder is an out 75% of the time. So, by comparing changeup whiff and grounder rates by the z-score method, and then weighting each rate accordingly (1 GB = 1.67 whiffs for the change), we can come up with an Arsenal Score for each changeup in baseball. 

For fun, here are the top 15 changeups thrown by starters last year by Arsenal Score. 

Top changeups

Now to find the changeup-ditchers, we need to flip the script. Let's look for the less established pitchers with the worst arsenal score on their changeups who also throw curveballs and sliders (plus cutters) regularly. (Madison Bumgarner's change ranked 13th-worst by Arsenal Score, but we're not arrogant enough to change what he's doing right now.)

Worst changeups

Judged solely by how often the pitchers in this group have been traded, you'll probably find this list includes young talent that inspires wide-ranging opinions. They're also relatively unformed, averaging 24 years old and 297 innings pitched so far. Here's an individual prognosis for each:

Shelby Miller

Changeup whiff rate: 2.9%

Changeup groundball rate: 38.6%

Curveball x, y movement: 6.6, -5.5

The league's regularly-thrown changeups had a 15.7% whiff rate and 49% groundball rate, so Miller's came up well short on both accounts. Miller actually threw the change second-least on this list --€“ he threw only 68 last year -- so maybe he's moving in that direction. It did have more arm-side run than average, but less drop, and the velocity gap wasn't really there. The average roundhouse curve, though, has 6/-6 movement from a righty, so Miller's is almost the prototypical 12-to-6er. He threw the cutter three times more in 2014 than he ever had before, and it got above-average results by whiffs. Perhaps all Miller needs to do is throw that cutter more and work on fastball command (up in the zone?) in order to re-find himself. 

Nathan Eovaldi

Changeup whiff rate: 5.3%

Changeup groundball rate: 38.5%

Curveball x, y movement: 5.4, -7.8

Like Miller, Eovaldi is in the process of shedding his change, perhaps. He threw only 57 of them last year. Last year, his curve got the best results of his career, too, with a 10.4% whiff rate that was finally average for the pitch. He's always had the velocity -- top five in fastball velocity this year -- and the slider is excellent. His change goes 10 mph slower than his fastball, but it doesn't have the arm-side run, the drop, or the whiffs associated with a good change. He might be the most Richardsian in this group.

Trevor Bauer

Changeup whiff rate: 7.1%

Changeup groundball rate: 40%

Curveball x, y movement: 5.3, -8.4

Bauer's change doesn't have great arm-side run or drop -- in fact, it has two inches less in both directions than the average major-league changeup from a righty. Though it has almost a 10-mph gap in velocity from his fastball, the change isn't getting whiffs. With an 18.6% whiff rate on his curve, Bauer's yakker was third-best among starters that threw 300+ curves last year. Since he already owns an average slider and good velocity on the fastball, Bauer has what he needs without the change. Perhaps he should throw it less often.

Vidal Nuno

Changeup whiff rate: 8.2%

Changeup groundball rate: 36.7%

Curveball x, y movement: -5.2, -5.3

Nuno doesn't do a lot of things well, but his change isn't good by any measure. It doesn't have a velocity gap (seven mph), and it has less fade and drop than his sinker. His slider (11% whiffs) and curve (9.4% whiffs) aren't great, but at least they are okay by benchmarks for those pitches. 

Zack Wheeler

Changeup whiff rate: 7.0%

Changeup groundball rate: 50%

Curveball x, y movement: 6.3, -8.6

Along with Bauer and Eovaldi, Wheeler may be the poster child for this mode of analysis. If you look at the rest of his arsenal, he has what it takes to succeed. His 14% whiff rate on the curve is top-20 for the pitch among starters, and that held up through 518 thrown. His slider is above-average by whiffs and grounders, and he has top-five fastball velocity among qualified starters. The only thing complicated matters is that Wheeler has great arm-side run on his changeup (nine inches, almost three inches more than average), and it gets a good amount of grounders. If he focuses on using the change more like a sinker, maybe it should remain in his toolbox. 

These pitchers are young enough that their pitching coaches might just want to continue trying to refine that change. Maybe another grip, another side session, another approach might help them figure it out. These pitchers also have good advisors around them. We don't need to tell them what to do. 

But it's still important to remember that the changeup is not the only way to a full arsenal. The curve offers starting pitchers a way out by busting platoon splits and changing speeds. Maybe a couple of these young pitchers will ditch the change and embrace the curve this year. Maybe it'll lead to their breakout performance.

share