/cdn.vox-cdn.com/uploads/chorus_image/image/65265411/1174723262.jpg.0.jpg)
The arc of human physical capacity always trends upward. It was revolutionary when Roger Bannister cracked the four-minute mile in 1954. Now, the record is nearly 17 seconds faster, held by Hicham El Guerrouj. The same is true of fastballs. Pitchers keep finding new ways to throw harder...except this year they didn’t.
Among pitchers with at least ten innings this year, average fastball velocity was 93.45 miles per hour (according to Pitch Info Solutions data available on FanGraphs). In 2018, the average was 93.48. That’s a drop of 0.03 mph. If it doesn’t seem like much to you, that’s because, uh, it isn’t.
Does this mean velocity as a whole has peaked? Not necessarily. There have a been a record 823 pitchers used across MLB this year, 623 of which have thrown at least ten innings. Over the final two weeks of the season, those numbers will tick upwards a little higher. The previous records of 799 total pitchers and 619 with ten innings were set last season. If teams are using more pitchers than ever, the quality dilutes.
An average decline of 0.03 mph is nothing to write home about, but that decline has not been universal. By putting every pitcher’s fastball into a percentile ranking, we can create velocity categories. Ignoring the outliers in the first and 100th percentiles, here’s how velocity has changed since last season:
:no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/19205894/Velocity_Change_from_2018_to_2019.png)
It makes sense that the bottom ten percent of pitchers declined by more than 0.2 mph, at least at first. The pitchers who wouldn’t have made the majors last year might account for the difference. The problem with this theory is that the low-velo guys mostly aren’t young or inexperienced. There’s a lot of Kyle Hendricks and Zach Davies types, as well as some Sergio Romo and Richard Bleier relievers.
It seems more pitchers are just getting away with throwing softer pitches. Look at Zack Greinke, for example. He averaged 90.0 mph in both seasons. In 2018, this ranked 557 out of 602, or 6.7th percentile. In 2019, he ranks 565 out of 615, or 7.6th percentile. It’s a subtle difference, but it indicates there are more pitchers are throwing softer than him this season.
After we get past the junk-balling bottom 10 percent, the 10-50 percentile are basically unchanged. 10-25 percentile increased by 0.03, and 25-50 percentile by 0.05. Both of these changes are likely too small to mean anything. In both years, this represents pitchers throwing 90.5-93.6 mph. If anything, it’s noteworthy that these below-average-velo pitchers remained stagnant while others dropped.
Perhaps the most interesting development is that the top half of velo decreased by about 0.1 mph. Logically, these pitchers should stay the same, or even increase velocity. This is difficult to explain, but here are a few hypotheses:
- Some pitchers haven’t yet thrown ten innings, thereby removing themselves from the sample in 2019. Cubs reliever James Norwood threw 11 innings in 2018 with a 97.9 mph fastball, which reached the 96.7th percentile. This year, he’s only thrown 5 1⁄3 innings, mostly as a September call-up. He might make it to ten innings before season’s end.
- With an increased emphasis on breaking pitches across baseball, teams might be less enamored with pure gas. The Pirates used Dovydas Neverauskas’ 97.3 mph, 93.8th percentile fastball for 27 innings last year. This year, he’s failed to reach ten innings, and his phone didn’t ring when rosters expanded.
- Along the same lines as #2, the existing hard-throwers might be more interested in movement. Perhaps some of them have added just a little cut to give their fastball some wiggle. This could come at the expense of a small amount of velocity.
- This is just a blip, and doesn’t really mean anything. We’re dealing with really small numbers here. We can’t make any definitive conclusions.
Whatever the reason, it’s fascinating that average fastball velocity is actually trending downward ever-so-slightly. It’s also interesting that there are different trends for different velo categories. We can only guess about the causes or meanings, because no one can be certain— not even Roger Bannister.
Daniel R. Epstein is an elementary special education teacher and president of the Somerset County Education Association. Tweets @depstein1983.