What’s up with Carlos Carrasco?

After Sunday night’s ugly 14-4 loss to the Los Angeles Dodgers dropping the Mets to just one game over .500 and 2.5-games behind the NL East-leading Atlanta Braves, the season seemingly hangs in the balance in mid-August.

Carlos Carrasco was roughed up to the tune of six runs on six hits in only 2.0 innings, and his ERA ballooned to 10.32 through four appearances with the Mets. A player who was supposed to be a huge lift to the rotation coming off of injury has been anything but – Carrasco has allowed 10 runs in 3.0 innings in his last two starts, and hasn’t lasted longer than 4.1 innings in any of his other appearances.

Coming back from a fairly severe injury, a torn hamstring, it is probably not unpredictable that the righty might struggle out of the gate. It is certainly unfair to him that the Mets failed to make a move at the deadline to acquire another starting pitcher outside of Rich Hill and pin their hopes on Carrasco’s return to stabilize a rotation that has been thrown into chaos by a rash of injuries and setbacks.

A look further into the numbers reveals something potentially concerning for the long-haul about his performance in a very small sample.

Examining publicly-available Statcast data at Baseball Savant, we see Carrasco’s pitch usage throughout his career has followed a predictable arc. His first few years, he was mainly a fastball pitcher (55.7% of his pitches were either four-seam fastballs or sinkers in 2015, the start of the Statcast era), but as fastball usage has plummeted around baseball, so has Carrasco’s.

In 2021, his most commonly thrown pitch has been his slider, used 30.5% of the time. His four-seam fastball rate is down to 27.1%, but his sinker is up to 21.5% after bottoming out at 4.8% in 2020. On paper that isn’t a bad idea. The difference in the underlying results between Carrasco’s four-seamer and his sinker has historically not been very large. Sinkers are a traditional pairing with a slider because of the divergence in movement, plus with Javy Baez and soon Francisco Lindor anchoring the up-the-middle defense, generating more groundballs isn’t a bad thing.

Carrasco’s biggest problem is that his slider is completely different from what it was in 2020, and that is leading to some interesting results. The pitch is still good enough to generate a 36.8% whiff rate, but from a velocity, spin, and shape it is wildly different.

Throughout his career, he has thrown a mid-80s gyro slider, characterized by more vertical drop and less horizontal movement. Carrasco has also had a high-spin slider, peaking at an average of 2,816 rpm in 2020.

This year, Carrasco still throws a gyro slider, but is throwing it harder than he ever has with an average velocity of 87.5 mph, up from 86 mph in 2020. This has led to a huge drop-off in the vertical break of the pitch – from an above-average 36.4 inches in 2020 to a below-average 31.3 inches in 2021. He has also only generated 1.8 inches of horizontal break, the least of his career on the offering.

His spin rate on the slider has fallen off a cliff to 2,355 rpm this year. That could be the result of a confluence of factors – when you throw it harder it spins less and moves less, a decrease seen because of changes made to the pitch, or simply MLB’s crackdown on sticky stuff. Unlike many pitchers we don’t have early season data to look at for Carrasco’s spin rates to see if they would have dropped when the new checking policies went into effect, but a change of nearly 500 rpm in one year is suspicious.

Consider the following two images. The first shows the spin-based and observed movement on Carrasco’s pitches in 2020 and the second the same factors in 2021.

Note the dramatic change in his slider. It is almost like he is throwing a completely different pitch this year.

Where this becomes problematic is that the xBA against his slider this year is .231, which seems like a low number but is 20 points higher than in 2020, and more than 30 points higher than in any other year of his career. The pitch that he throws most often has been less effective this year, and that is a problem he and pitching coach Jeremy Hefner will have to solve for the Mets to stabilize their rotation.

Joe Vasile is a broadcaster for the Scranton/Wilkes-Barre RailRiders (NYY, AAA) and the host of the baseball history podcast ‘Secondary Lead’.