Can data analytics resurrect Matt Harvey?

On February 13, an alert popped up on my phone telling me that the Orioles had signed Matt Harvey to a minor-league deal. I, like other Orioles fans, I imagine, got a thrill of recognition at the name. What? The Dark Knight, an Oriole? Then I thought, When’s the last time I actually saw Harvey pitch? So I checked out recent footage to see what was what…

Blech. Here is 2019 Matt Harvey getting eaten alive by the Twins during a start for the Angels. (Note the first guy to go yard off of him: Jonathan Schoop.) One flat frisbee of a fastball, badly located, and not much break on the off-speed stuff. It reminded me of Chris Tillman right before he retired.

I concluded the whole signing was a gimmick.

Then I came across a longish piece by the Baltimore Sun on Harvey’s “offseason transformation.” The one-time ace, fed up with five-plus years of poor results after Tommy John surgery in 2013 and thoracic outlet surgery in 2016, had visited one Baseball Performance Center (BPC) in Pleasantville, NJ, and worked out for four days with trainers who use data and imaging to develop pitchers. On February 10^th, BPC started an aggressive social media campaign hyping the new-and-improved Harvey. Three days later, the Orioles signed the righty to a minor-league deal. Coincidence? Either way, it was definitely a good story, the one-time ace harnessing data to rediscover his sky-high potential. So . . . these wizards who bring back arms from the dead, could their dark arts help Matt Harvey?

After a moderately deep dive into Harvey’s delivery, stats, and prognosis in trying to bounce back from Thoracic Outlet Syndrome surgery, the answer is … definitely maybe. OK, here’s a better answer: the 31-year-old Harvey is still capable of lighting up a radar gun, and even more intriguingly, there’s evidence that if he’s harnessing his new lessons, the results might be competitive. Which makes the big question: if a former ace literally has to relearn to pitch, can a fan expect ace-like consistency, not to say results, from him at all?

It’s weird but important to remember that, eight seasons into his career, Harvey is by no means a “finished product.” The fastballing stud had shocking success back with the Mets when he threw an easy 100, putting up a 2.27 ERA with a sub-1 WHIP and 191 K’s in 178 innings in 2013. (Here’s a flashback composite by one nostalgic Mets fan; there are lots of these—Harvey flashback composites and nostalgic Mets fans. It’s quite a sight.) But apparently, there was something seriously wrong with the delivery, and soon injuries robbed Harvey of his success. After a 2017 season where his ERA ballooned to 6.70, the Mets dealt him to Cincinnati in 2018, where he finished a mediocre 7-7 with a 4.50 ERA. Free agency took him to LAA in 2019, a stint about which the less is said, the better. (An LAA fan tastefully put up a black screen and captioned it, “Matt Harvey’s Angels highlights.”) One 0-3, 11.57-ERA season in seven games for Kansas City later, and you understand how the one-time Dark Knight landed with us on a minor-league deal.

Harvey claims that injuries robbed him of his feel for pitching, and that he’s had to relearn his delivery to get it back. “You get used to not really having to pay attention to that stuff [mechanics],” Harvey explained in a thoughtful, candid interview when he arrived in Baltimore in February. “The first couple years [of my career] I threw so well that it just came easy, came natural.” Obviously, four days at an offseason clinic weren’t going to fix five or so years of bad habits, and reportedly, a big reason Harvey signed with the O’s was their analytics regime. Right after he arrived, pitching coaches Darren Holmes and Chris Holt went to work on him. With reporters cut off from access to the team, and MASN churlishly broadcasting two (maybe) Orioles games this spring training, it’s been tough to get a sense of what Holt and Holmes may have worked on with Harvey. But the one-time ace seems to like what they’ve been telling him. As he told reporters after his encouraging start on Monday:

[I]nstead of me searching and trying to figure out what I am doing wrong and how to fix it, [Holmes and Holt have] been able to pinpoint that and kind of show me videos from when I threw before. [W]hether it’s load your back hip a little bit more or you know, create some scapular flexion here, look you used to do this and you’re not doing it anymore … they’ve been unbelievable with that and obviously the work’s not done. We’re not satisfied and we’re going to keep pounding those mechanics and the work in so that everything can fall into place and just concentrate on getting people out and executing each pitch.

So far, in six innings pitched, Harvey has given up nine hits and five runs, for a 7.50 ERA with four strikeouts. I know, I know, don’t lose your hat. But in Harvey’s last start, he gave up just a run in four innings, touched 96 on the radar gun, and struck out three. If the problem isn’t a dead arm so much as poor mechanics, is a revival possible?

If you take BPC’s word for it, there were important and obvious things to fix in Harvey’s delivery.

On the left is 2020 Matt Harvey with Kansas City; compared to his 2015 self, he is wide open, his fingers unable to get behind the ball, and therefore to produce spin.

Then there are the stats. Not just MPH and XBA, but the stats that make experts the big bucks: spin rate, horizontal & vertical break, release point, pronation, etc. According to BPC, Harvey showed up to them with a four-seam fastball that had ticked up in velocity to 93-97 mph since a low in 2018 but had no “carry,” i.e. no vertical break. This meant Harvey couldn’t target the top of the zone with it anymore, as he used to, because the pitch was just a flat meatball. (I am, of course, paraphrasing.)

Looking at this data on Harvey’s fastball suggests that BPC might not just be whistlin’ Dixie. Harvey’s fastball didn’t lose much velocity after Tommy John surgery in 2013, and since his TOS surgery in 2016, it’s ticked up from a low of 93.2 MPH on average to 94.0 in 2020. What has consistently trended downward is whiff% and fastball spin (from 2335 rpm to 2139 rpm, a seemingly significant drop).

One more take on this: here is the vertical movement on Harvey’s fastball compared to average. Harvey’s four-seamer never had a ton of sink, even in the glory days, but the tendency has gotten worse in later years.

A final interesting issue is Harvey’s thoracic outlet syndrome: the news ain’t all bad. For one thing, the disease, which causes tingling and loss of sensation in the fingers, helps to explain his loss of command and decreased spin rates. On the other hand, a recent study of 27 MLB pitchers who underwent TOS surgery found that 74% were able to return to play at the MLB level, and with metrics that were mostly unchanged from prior to surgery.

So, the question is, can Harvey repeat the lessons he’s getting on spin rate, spin axis, release point, and so on? He seems to think it’s getting easier. As he said on Monday, “You know, the idea of pitching is to focus on the attack and to go after each pitch. [I]t’s hard to do that when you’re thinking about five different things mechanically. Today was a huge step in the right direction where I was actually able to, for the most part, just concentrate on attacking the zone and not getting caught up in mechanical things.”

As The Sun’s Jon Meoli wrote on Tuesday, Harvey’s progress is happening “in a tunnel.” So it’s really tough to say what’ll happen. For now, some sneaky footage we have of him looks, actually not half-bad.

Bottom line: this is not a pitcher who’s completely out to pasture. Harvey is a guy who always had pitching come easy to him, lost it, and now is trying to process and apply advice on how to pitch in a results-oriented way. Does he have a comeback in him? It’ll be fantastic, if it works.