Fastballs with Diverging Spin Rates and Velocity

The inspiration for this study came from this blurb in a recent Mining the News.

In the past, the general rule is that a higher fastball velocity leads to more strikeouts. Once spin data became available, the link between more spin and strikeouts was quickly found.

And they can go in separate directions. After going way too deep down this rabbit hole, I came out with a simple formula.

Before diving in, a few items need to be cleared up. First, other factors besides velocity and spin can affect a fastball’s effectiveness such as pitch location, deceptiveness out of the hand, and how often the batter has seen the pitch. The key with this study is to have a simple formula to start evaluating fastball changes for the accessible stats.

Second, I tried to use Bauer Units (simply spin/velocity) to determine how a fastball performance would change. While there was some correlation between Bauer Units and pitch performance, it became unnecessary noise. Seperating out spin and velocity was a better approach. Bauer Units are more helpful with pitch design than pitch evaluation.

Now, to the analysis. I found all the pitchers from 2015 to 2020 who threw 200 four-seamer fastballs in matched seasons (n=1134). Sinkers were not included because the goal with them is to remove spin and get the ball down through the zone.

After looking at several possibilities, I ended up with the following three rules of thumb for determining a change in a fastball’s swinging-strike rate. The r-squares aren’t high for any of the values (~.1 to .2), but that’s expected with so many inputs into a fastball’s performance.

Note: Remember that spin and velocity are related, don’t combine the first two formulas. Use the third one.

Spin: 100 rpm change * 1.0% SwStr%
Velocity: 1 mph change * 0.9% SwStr%
Spin & Velocity: 1 mph change * 0.78% + 100 rpm * 0.5%

Going back to the Paddack example (+0.2 mph, -60 rpm, -2.0 SwStr%), here are the expected changes.

Spin only: (-60/100) *1.0% = -0.6%
Velocity only: 0.2 * 0.9% = +0.2%
Spin & Velocity: 0.2 * 0.78 + (-60/100) * 0.5% = -0.14%

The loss of spin outweighed the velocity loss but the entire decline in swinging-strike rate can’t be explained with just the two factors. Paddack and the team identified those other causes and hope to correct them.

That’s it for today. I believe Mining the News will be more important than going over a few more examples. The formula will be useful once the regular season rolls around and spin and velocity data are available on every pitch.

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 three FSWA Awards including on for his MASH series. In his first two seasons in Tout Wars, he's won the H2H league and mixed auction league. Follow him on Twitter @jeffwzimmerman.

newest oldest most voted

Spin axis is obviously hugely important, and while the actual physics is what they call “complicated” maybe there’s a quick-and-dirty weighting factor ( eg something like k*Cos(spinaxis) ) that can be applied to the spin rate to give you a better fit?