Baseball Prospectus is looking for a Public Data Services Director. Read the description here.

Matt Harvey dominated this weekend, though instead of an opposing lineup, it was the news cycle.

A quick recap, for those of you who spent the weekend in blessed solitude with little or no access to Twitter: Matt Harvey has thrown 166 1/3 innings so far this season. According to the Mets, Harvey had a “soft cap” of 180-185 innings, and was asked if he was comfortable going to 190-195 innings. According to Harvey’s agent, Scott Boras, the pitcher was given a “hard cap” of 180 innings to throw this season by the orthopedic surgeon James Andrews, a number that might have been decided after the Mets asked Harvey to throw more innings in early August. Finally, Sunday afternoon, Harvey published his own account in The Players Tribune, promising (with few specifics as to how) that he would pitch in October.

In pitch terms, Harvey has thrown 2,459 pitches in major-league games this season. Is that a lot? It it too many? Is it enough to exhaust a pitcher? Have those pitches already affected Harvey? His pitch total is about 250 pitches lower than that of Max Scherzer, who has thrown 184 innings, and it’s more than 500 behind David Price, who has thrown the most pitches in baseball—but, of course, neither is coming back from a missed year or a major surgery. There aren’t any pitchers coming back in exactly the same situation as Harvey, so there isn’t a great like-to-like comparison to be found. However, one way to tease some information out of his performance is to compare Harvey to himself.

Harvey’s average velocity has dropped since its season high in May, as seen here:

On its own this isn’t necessarily worrisome. Pitchers experience small changes in average velocity across a season regularly, and such spikes and falls can often be attributed to common factors such as rest and experience. However, before his first Tommy John surgery, Harvey experienced a similar spike and drop off across June 2013 to August 2013.

Perhaps more notable than the relatively small changes to his monthly averages, Harvey lost miles per hour off his fastball in the later innings of his September 2nd start. While that could be an indicator to keep an eye on, this was also the day where he was treated for dehydration and weakness after the game, and he didn’t fly with the team to Miami.

Though there’s minimal research on how pitch selection might be tied to mechanical comfort, or whether changes in pitch selection can have reliable predictive value, it is worth looking to see whether a pitcher has dramatically changed his usage across time, particularly if these changes correlate to other changes in recordable data. The pitch selection change of largest note occurred in July, where Harvey upped his sinker usage by 10 percentage points, jumping from 1.9 percent in June to 12.4 percent in July[1].




































If the sinker spike came at the expense of four-seamers, it probably wouldn’t mean much, but of more interest is Harvey’s gradual disinclination to use his slider—a pitch that has shown some correlation to elbow pain, and that some cautious pitchers have moved away from—despite its supposed resurgence July. A general look at batter handedness shows that this decline is not just because he faced fewer batters of one hand, but that he’s used the slider less across the board as the season’s gone on. In September, his slider usage has dropped to 8.9 percent, though that’s an incomplete sample thus far. Additionally, Harvey has grooved his slider more on average than he did in 2013, according to Brooks Baseball’s pitch categorizations and he’s gotten fewer swings on the slider this year, from 53.4 percent to 45.9 percent. (Though, interestingly, a higher percentage of whiffs, from 16.5 percent to 18.2 percent.)

There’s also no real indication that Harvey’s mechanics or release point have changed dramatically across the season, with any variation in assumed release point staying within a few inches, vertically. Now, this isn’t to say that there hasn’t been some subtle change in mechanics, or that there wouldn’t be a change in mechanics if Harvey pitches with fatigue, but there’s nothing clear in the data right now. Likewise, there hasn’t been a large difference in the movement that Harvey is getting on his pitches across 2015, other than a slight drop in all movement in, when else, July.

While Harvey is likely experiencing some fatigue due to pitching a heavy workload after a full year off from pitching, there hasn’t been a significant effect on the data gathered by PITCHf/x—only strands at which we can pull without knowing which matter. None of the ominous, flashing warning signs that we tend to expect to show up in this kind of data are apparent, and while the pitcher himself is the best judge of his body, it’s also easy to understand why the Mets would see no issue with trying to get more innings out of Harvey’s exceptional arm.

[1] There is some margin for error with Harvey’s SI compared to his FA. The pitches are just different enough to say that there was a jump of some sort in July, and there is enough confidence in the data tagging to say that it was a significant jump, but there is a slight need for a moderate grain of salt, as the two can blend together annoyingly.

[2] I’ve skipped April in this table, as that was his first month back from Tommy John and the numbers may have been less reliable than May/June/July/August.

[3] The Mets skipped one of Harvey’s August starts, hence the lower number of pitches across this month.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
What source are you using for your data, raw Pitch F/x? In any case, there's clearly a pitch classification issue going on with Harvey, as evidenced by all of his sub-92 mph fastballs suddenly being called non-FBs 10 starts ago:

With such inconsistent data, I would hesitate to draw even mild conclusions about his in-season velocity change. The good thing is that his max velo has been relatively stable all season (97-98). Plus, his post-AS Break xFIP has been 3.23, vs. 3.44 in the first half.
Kate, like all BP authors, has access to the Pitch Info data that I provide to Brooks and several MLB teams.
Not sure I follow. What data are you using as your source? Do you classify pitches with your own algorithm? Finally, is your data available on this site?
I've been classifying pitches since they started publishing data. The data is used on
And, no, the data is not available for download.
*it's used on the leaderboards, also byproducts of the data are used in our stats, particularly framing; the process is not automated, it's manual. And we have a method to recalibrate the data so it's more consistent park-to-park

Thanks, Kate.
As I recall, Stephen Strasburg was a similar case in 2012. The PITCHf/x data showed nothing indicating obvious fatigue, so if it was there it was too subtle to measure that way.
There was a lot of discussion on SNY during Harvey's recent starts (particularly, IIRC, his 8/11 start against Colorado) that he had decided to start dialing back a bit and pitching more to contact so that he could last later into games. That CO start is particularly out of line with vintage Harvey--8 IP, 0 R, 4 H, and only 4 K. So that could be a big reason his velocity is down across the board in August. The change in pitch selection is a little more worrisome though.