Velocity Changes During the 2018 Season
It may seem that fastball velocity gets over utilized for explaining a pitcher performance, but I don’t think it gets used enough. As I get going full steam ahead in my off-season research, I find myself always looking to see how a pitcher’s velocity held up over the season. Instead of looking up each pitcher individually, I decided to go ahead and collect 2018 fastball velocity reading for an easy reference.
For the data, I found the pitcher’s velocity for the whole season, April to June, July to October, and just September/October. The entire data dataset can be found here.
I went through the data and found some intriguing pitchers who gained and lost a few ticks. I divided up the analysis between first versus second half and first half and September. Many of the same pitchers would have made both lists. Here is the first to second half values.
| 2Hv-1Hv | Pitcher | Fastball | Count | Season Velo | 1H Velo | 2H Velo | Sept Velo |
|---|---|---|---|---|---|---|---|
| -1.7 | Stephen Strasburg | FF | 996 | 94.6 | 95.3 | 93.6 | 93.3 |
| -1.2 | Dereck Rodriguez | FF | 710 | 91.6 | 92.4 | 91.2 | 90.9 |
| -1.1 | Marco Estrada | FF | 1230 | 88.5 | 89.0 | 87.8 | 88.0 |
| -1.0 | Antonio Senzatela | FF | 976 | 93.6 | 94.4 | 93.4 | 93.1 |
| -0.9 | Tyler Skaggs | FF | 931 | 91.5 | 91.8 | 90.9 | 90.0 |
| -0.9 | Brad Keller | FF | 992 | 94.3 | 94.9 | 94.0 | 94.3 |
| -0.8 | Marco Gonzales | SI | 650 | 90.1 | 90.4 | 89.6 | 89.3 |
| -0.8 | Jon Gray | FF | 1387 | 94.7 | 95.0 | 94.3 | 94.2 |
| 1.1 | Zack Wheeler | FF | 1295 | 95.8 | 95.3 | 96.4 | 95.8 |
| 1.1 | Mike Clevinger | FF | 1735 | 93.6 | 93.0 | 94.2 | 94.5 |
| 1.2 | Gerrit Cole | FF | 1617 | 96.2 | 95.6 | 96.7 | 96.5 |
| 1.2 | Jacob deGrom | FF | 1399 | 96.0 | 95.4 | 96.7 | 96.9 |
| 1.3 | Jordan Hicks | SI | 707 | 100.5 | 99.7 | 101.0 | 101.1 |
| 1.4 | Cole Hamels | FT | 450 | 91.4 | 90.8 | 92.2 | 91.8 |
| 1.5 | Gio Gonzalez | FF | 885 | 89.8 | 89.0 | 90.5 | 90.3 |
| 1.5 | Lucas Giolito | FF | 1305 | 92.3 | 91.7 | 93.2 | 92.5 |
| 2.2 | Matthew Boyd | FF | 904 | 90.4 | 89.3 | 91.5 | 92.4 |
Stephen Strasburg is the most concerning name on the list with his fastball down ~2 mph. The drop occurred after he went on the DL with shoulder inflammation.
His results also took a hit with his ERA going from 3.46 to 4.20 and his K%-BB% dropping from 23% to 19%.
Strasburg is getting in the danger zone where his fastball will start losing its effectiveness if he loses any more velocity. Here are its swinging-strike rates since he joined the league at different velocities.
| mph | SwStr% |
|---|---|
| 92 | 4.8% |
| 93 | 7.0% |
| 94 | 12.7% |
| 95 | 13.9% |
| 96 | 16.8% |
| 97 | 14.7% |
| 98 | 16.0% |
The fastball starts to lose its effectiveness as it dips near 93 mph.
With the velocity drop, Strasburg will still be a good pitcher because his changeup is elite and curve and slider are decent. His issue is that he’s already been cutting his fastball usage from 73% when he entered the league to 52% last season. I think the chances of him having that elite season has passed.
One major consideration will be if he can get his walks under control. In the first half, they were at 2.1 BB/9 but jumped to 3.5 BB/9 in the second half. Spring training reports are going to matter quite a bit on how he gets valued.
| SEPv-1Hv | Pitcher | Fastball | Count | Season Velo | 1H Velo | 2H Velo | Sept Velo |
|---|---|---|---|---|---|---|---|
| -2.2 | Danny Duffy | FF | 1044 | 93.0 | 92.9 | 93.1 | 90.7 |
| -2.1 | Chris Sale | FF | 957 | 95.3 | 95.1 | 95.9 | 93.0 |
| -1.6 | Corey Kluber | FC | 922 | 88.5 | 88.7 | 88.3 | 87.0 |
| -1.5 | Aroldis Chapman | FF | 679 | 98.9 | 99.1 | 98.5 | 97.6 |
| -1.3 | Bartolo Colon | FT | 1369 | 86.9 | 86.9 | 86.8 | 85.6 |
| 1.4 | Luis Castillo | FF | 944 | 95.9 | 95.5 | 96.4 | 96.9 |
| 1.5 | Patrick Corbin | FF | 630 | 90.8 | 90.5 | 91.1 | 92.0 |
| 1.6 | Tyson Ross | FF | 1226 | 90.7 | 90.4 | 91.1 | 92.1 |
| 1.8 | Drew Pomeranz | FF | 576 | 89.4 | 88.9 | 89.9 | 90.8 |
| 1.9 | Sean Newcomb | FF | 1833 | 92.9 | 92.7 | 93.3 | 94.5 |
| 2.0 | Mike Fiers | FF | 1095 | 89.4 | 88.5 | 90.2 | 90.5 |
| 2.7 | Martin Perez | FT | 703 | 92.5 | 91.4 | 92.9 | 94.1 |
Corey Kluber is the name which jumps off. For pitcher going in the second round of mock drafts, an ~1.5 mph drop in his fastball throws a major red flag.
He got hit around more in the second half with his BABIP jumping from .248 to .321 while list strikeout and walk rates remained constant (22.5 K%-BB% to 22.1 K%-BB%). Kluber found a way to be effective even with the velocity loss.
My gut says something is off but I can’t find it. His steamer projection has his ERA next season back in the 3.50 range (same as 2015). so it even sees that something is not right. I’ll read some more previews as the season nears and see if I can gain a better understanding of him.
Jeff, one of the authors of the fantasy baseball guide,The Process, writes for RotoGraphs, The Hardball Times, Rotowire, Baseball America, and BaseballHQ. He has been nominated for two SABR Analytics Research Award for Contemporary Analysis and won it in 2013 in tandem with Bill Petti. He has won four FSWA Awards including on for his Mining the News series. He's won Tout Wars three times, LABR twice, and got his first NFBC Main Event win in 2021. Follow him on Twitter @jeffwzimmerman.
Good scratching of the surface. Some things I’d love to know: Relationship between loss of velocity from 1H to 2H and change in RA9 or FIP. Is there a linear relationship or does it depend on starting velocity? What about next year? What happens next year when pitchers lose or gain velocity from 1H to 2H (in velocity and RAo9 or FIP)? And what happens to pitchers next year when lose or gain velocity only in September ? Next article?
You should be his editor! I want all of these questions answered and will gladly click and read if there is a follow up!!
Most of your question can be answered in this article by Mike Fast:
http://www.hardballtimes.com/lose-a-tick-gain-a-tick/
And I relooked at later:
https://www.fangraphs.com/blogs/instagraphs/revisiting-mike-fasts-gain-a-tick-lose-a-tick-study/
As for how much to weight each one, that’s a great idea and I’ll look into it next article.