Jharel Cotton’s Changeup Is Objectively Impressive

Oakland right-hander Jharel Cotton made his major-league debut on Wednesday. The results were positive: over 6.1 innings, Cotton conceded just a lone run on two hits — the product, that run, of a homer by the Angels’ C.J. Cron. The process, while entirely adequate, was also slightly less positive: over those 6.1 innings and against those same 22 batters, Cotton recorded just three strikeouts.

In a sense, this start was the opposite of the sort which have defined much of Cotton’s season in the Pacific Coast League this year. Despite producing the best strikeout- and walk-rate differential (K-BB%) among all 57 Triple-A qualifiers in 2016, Cotton also recorded a 4.31 ERA — which, it turns out, is only the 35th-best ERA at Triple-A and even pretty middling among just PCL starters, too. The home runs were a problem for Cotton. Sequencing was a problem for Cotton. Controlling the strike zone wasn’t.

Apart from the runs he allowed and the runs he might have been expected to allow — whatever the discrepancy there — Cotton exhibited one quality yesterday that he’s exhibited all of this season and all of last season and maybe always since he was just a small child. An excellent changeup, is what. Lead prospect analyst Eric Longenhagen, among others, has described it as a plus-plus pitch — and it’s the presence of that pitch that has largely been credited with allowing Cotton to experience such great success as a professional despite a rather diminutive frame.

On Wednesday, Cotton threw the changeup approximately 25% of the time — a rate of usage that would place him fifth among the league’s 79 qualifiers by that measure. It only led to four swings and misses (a rate of 16.0%, only slightly above league average), but it was Cotton’s best offering of the day in terms of runs prevented per pitch.

This seems like a reasonable moment to provide some video footage of the pitch in question. By way of MLB.com, here are the final pitches from two of Cotton’s three strikeouts, both by way of the change:

So, here’s something that we know: that Cotton’s changeup is regarded as objectively excellent. And here’s another thing: he threw it a lot during his major-league debut. And a last thing: it was probably his best pitch during that debut. We pretty much know these things. But here’s something we don’t know: how does Cotton’s changeup compare in objective terms to other pitchers’ changeups?

There’s a lot of ways to answer the question, probably. The way I’ll use is to look at the separation of velocity and movement of Cotton’s changeup from his fastball. Why I’m doing that is because, last year, FanGraphs contributor and arrant fool Eno Sarris endeavored to find the best pitches by those variables, thus allowing me to stand on the shoulders of that weirdly coiffed giant.

What Sarris did was to find the differential in velocity between every pitcher’s fastball and changeup and also the differential in horizontal movement and also the differential in vertical movement. Assinging z-scores to each value, he was able to identify the pitchers who featured the greatest total differential — the changiest changeups, as it were.

Below is a modified version of a table he included in that post — in this case, featuring the top right-handed starters by the methodology he employed.

There are a lot of numbers here, but the most important of them are the three which appear in the lower right-hand corner of the table. Here we find the average differentials among the top changeups by velocity, horizontal movement, and vertical movement. So, for example, we learn that the pitchers in this group threw their changeups 8.4 mph slower, on average, than their fastballs. And 2.9 additional inches of arm-side run. And with 5.4 extra inches of “drop.” Again, all relative to the fastball.

How Cotton’s changeup fits in the context of those top changeups, is what the following table is designed to illustrate. For both his fastball and changeup, I’ve included (care of Brooks Baseball) both Cotton’s average velocity and movement from Wednesday’s start. I’ve then calculated the differential for each variable. Finally, I’ve reproduced the corresponding numbers from 2015’s best changeups above.

What do we find? Unsurprisingly, that Cotton’s changeup compares well — at least by this methodology. While the top changeups averaged about an 8 mph differential, Cotton’s was closer to 15 mph on Wednesday. Much slower, in other words. As for horizontal movement, Cotton’s is more understated, featuring a slightly lower differential than the average top changeup. So his pitch is less defined by arm-side fade, is what that means. Finally, there’s the depth of the pitch. The changiest changeups featured an average differential of 5.4 inches from the changeup. Cotton? Two inches more drop than that. It’s all quite promising.

Naturally, this methodology omits any consideration of mechanics. If a pitcher’s changeup doesn’t resemble the fastball out of the hand, then surely much of the velocity and movement differential is rendered moot. What one finds here, however, is that Cotton’s changeup is very likely one of the most extreme sorts featured by a starting pitcher in the majors.