Case Study– Good and Bad Discoball: Post 55

Michael Augustine (https://twitter.com/AugustineMLB) and I were recently discussing a couple of Strasburg “Discoball” changeups. The impetus of the discussion was Trip Somer’s new tool (Trip is @one_pitch_man) that allows one to enter tilt, efficiency and orientation of the pitch and visualize what that looks like spinning at any RPM from the pitcher or catcher perspective. I am having a lot of fun with that.

I used a clip of a particularly good pitch and he responded with a better quality clip on one that, to my eye, was not nearly as good. I am focused on orientation while Michael has the better eye for tilt and efficiency and what camera views do to our perception of them.

Here is my goto clip for a great “discoball”

And a better shot of a not as good changeup:

And Michael’s overlay of the two

The overlay may make the release of these two pitches look more similar than they are, so I am not basing many conclusions on the overlay itself. But what it appears to say is very much inline with my point here. One of these pitches is better than the other. The trackman data show that the release point was identical but that he released the bad one somewhat later.

My claim has been that Stras uses the top seam on the front of the ball to tilt the wake up and push the ball down. Michael eyeballs his change at 2:30 tilt and around 94% efficiency.

Here is what that pitch looks like with a good seam orientation from the pitcher’s view. I claim that the first game video (July 20, 2018) looks like this.

Looking at the same pitch from the catcher’s view, one can see a seam on the top front of the ball about 3/4 of the time. That will cause separation, and its ability to do so will improve as the ball drops and the hemisphere plane tilts forward.

Turning now to the bad, from that second game clip, I see the orientation of the ball having a larger top rotation. Note that I had that dialed in to -17 degrees for the good pitch, and the bad one look like -32, Compare the video below to the game video from 2019 and see if you agree that the ball looks like this:

If you are confused by the top orientation numbers, again, I refer you to our Post 51. This is central to this entire discussion. Tilt and Gyro are Magnus Effect, Orientation and Gyro are Seam Shifted Wake effect.

From the catcher’s view, now you can see that while that top seam is sometimes in the position we want, it is there much less often than for the good pitch.

In my opinion, Stras is not perfectly consistent with that orientation, and that is why his downward acceleration on his changeup varies so much. The histogram below shows the downward acceleration of all his 2019 changeups (red) compared to his fastballs (blue). These are normalized by the acceleration due to gravity, so the units are g’s. So -1g means the ball falls like a stone, less than -1g means the ball is being forced downward.

The takeaways from this plot are that his changeup has a broad distribution of downward accelerations and that the tail extends below -1g. Each of these are due to seam effects. Note his fastball has no seam effects.