Our high school owns a Hack Attack pitching machine. These machines have 3 solid wheels. We own the huge one that can do nearly anything. BUT, when you are operating it, any adjustment to the knobs alters RPM and velo at the same time, and often in the opposite way that a pitcher does.
For reference, this machine allows the user to adjust the speed of each of the 3 wheels independently with three knobs. The knobs have numbers with arbitrary units. For a fastball (which has backspin), the lower wheel must be spinning faster than the upper wheels. Curve balls are the opposite. If you want a faster pitch and increase the upper wheels to larger values, you inadvertently decrease the RPM.
I found that we were inadvertently choosing settings that resulted in RPM values our hitters will never see. To find the exact nature of the relationship between these knobs and the values we care about (RPM and velo), I set up a test.
The machine was in our batting cage along with a Pocket Radar Smart Coach and my Sony RX100 V high speed camera.
A flip scoreboard was used so we could match the high speed video to still images of the knobs and a radar reading.
The least glamorous part of this was counting the frames per revolution of the ball in the high speed videos and entering these plus the knob settings and velocity into a google sheet.
I hypothesized that the velocity is proportional to the sum of all the knobs and the RPM was proportional to the difference between the settings on the top and bottom. Amazingly, this proved to be exactly right.
Note that the sum of knobs is 2XLeft + Bottom since the left and right knob are assumed to be the same.
I denoted the slope and intercept of the velocity data as C1 and C2, respectively:
C1 = 4 and C2 = 6.5
And using D1 and D2 for RPM:
D1 = 832 and D2 = -250
The result of all of this is the following calculation for the knob settings for a given desired velocity and RPM. Note that this is only for a pitch with no side spin. In other words, I’m assuming the two top knobs have the same setting.
Bottom = 1/3 X [ (Velo – C2)/C1 + 2[(D2-RPM)/D1]]
Top = (RPM – D2)D1 + Bottom
This can easily be put in a google sheet and computed from your phone on the spot. I have entered it in to the calculation tab on the data sheet.
The usual cautions about extrapolation apply. If you want some side spin, I’d equally add to one of the upper knobs and subtract the same amount from the other knob.
If you use this and find it good or bad, I’d love to hear about it. Hit me up at @TheMagnusPI on Twitter.