Nearly everyday, I hear someone say they are “humbled” due to some recognition, award, promotion, etc. It sounds like BS, and I think it’s because it is BS. If you want to feel humbled, put a theory out there on how something works, but continue to ask questions and entertain the possibility that it is totally wrong, leading to the possibility that you learn it was totally wrong. Then you will know humility. I had this experience this week.
I continue to be convinced that one can throw a pitch that moves in a way that relies on an asymmetric wake to create force other than the Magnus effect. But, I am no longer convinced that the asymmetry is due to a laminar flow on one side and turbulent on the other.
My post titled “A Complete Description of the Laminar Express” contained an important error about the orientation of the ball. I mistakenly thought that a right hander throws this pitch with gyro that causes the axis on the first base side of the ball to be closer to home plate than the third base side. I was contacted by an anonymous reader (who asked to be called “Equity Private,” Cubs fan and pseudonymous proprietrix of www.finemrespice.com) who pointed out my error (very gently).
If the ball were oriented that way, it seemed possible that the disturbances caused by the seam on the 1B side would be damped out due to the forgiving pressure hill on that side, while the seam on the 3B side would cause transition to turbulence. It was a great theory, but the ball does not fly like that. It flies as pictured below, viewed from above and thrown by a right hander. I’ve sketched about 30 degrees of gyro spin.
My new explanation (suggested to me by my wife, who has little interest in baseball but was sitting next to me as I cursed my mistake) relies on one fact that we proved in an earlier study: That the boundary layer separation point (where the wake starts) tends to occur on a seam. The locations of boundary layer separation are marked with blue dots in the animation in Figure 2 of a 4-seam fastball moving right to left. Note how the separation location can be well past the hemisphere line when the seam is located there.
What I have sketched in Figure 1 shows the seams on the front of the ball having no effect (or, perhaps causing boundary layer separation in front of the hemisphere line), but the seam on the rear on the 3B side is forcing separation from the ball at a point beyond the hemisphere line. The other seam on the rear of the ball is well inside the wake and therefore does nothing. Since the ball is rotating, most of the time there will be a seam just past the hemisphere on the 3rd-base side. The amount of gyro that the ball has directly affects the location of that 3rd-base-side seam and should alter the asymmetry of the wake, up to the point where the boundary layer separates before the seam.
The PIV data we acquired with Driveline still tells the same story. The wake is asymmetric in a way that is consistent with Figure 1.
Conversely, if a 2-seam is thrown without gyro, the wake is symmetric.
Here is an excellent example of an MLB pitch, also sent to me by Equity Private.
It’s tempting to focus on the seam that sits near the vertical center of the ball, but that is doing nothing. The seam causing the movement is on the right side of the ball and harder to see.
I see the relationship between the gyro angle and the magnitude of the sideways force as looking something like this:
For small gyro angles, the seam is too deep in the wake to have an effect. As gyro angle increases, the seam approaches the hemisphere of the ball. The boundary layer separation point will snap to the seam location at some angle. The green and blue curves represent two possibilities. There are many since we don’t know how far to the rear of the ball the flow can stay attached. If the gyro angle increases more, the seam (and thus the separation point ) will move toward the hemisphere and the effect will diminish, completely disappearing near 45 degrees.
By contrast, check out this famous Freddy Garcia split finger fastball (taken from http://baseball.physics.illinois.edu/Garcia1.html). The thing to note is that, compared to the Kluber pitch, the left seam is closer to the hemisphere line. And the ball breaks sharply left, suggesting that the wake forms off the left seam instead of the right seam, as in Kluber’s pitch. This suggests that the “effect” in Figure 5 should probably be able to go negative.
So, I am arguing that transition from laminar to turbulence is not important. The smooth patch that everyone points to is not important. From everything we have seen in the lab, the flow on a baseball is not laminar even as slow as 70 mph, so I very much doubt it can be at 95 mph, where some claim to throw this pitch. What is important is the location of the seam on the rear of the ball.
Here are some things to try:
If you scuff the smooth spot, does it still work? (it should if I am right)
Does gyro angle correlate to pitch movement? (it should)
A new PIV study showing that the boundary layer separation point does lie on the seam consistently.
More examples of the pitch:
A more recent one from Jake Arrieta. Who knew he throws one?
So, if you find yourself convinced, maybe you will agree that the pitch needs a new name. I’d love to hear suggestions. I remain open to other ideas, and I remain not real certain about this or many other things!