Maybe it’s more of a hypothesis. I’ve been going on about this for a couple of weeks, but I’m not sure I’ve been able to make several of my points clear, so I’ll give it a shot here. Before diving in, note that I have changed the way I plot the data. Colors now indicate the velocity magnitude of the air, with Red being the ball speed and Blue being zero.
Also, in order to understand a lot of this discussion, you first have to understand “Flow Separation” or “Boundary Layer Separation.” Basically, the air on the front of the ball flows along the surface of the ball, but at some point on the back, it “separates” and forms the wake. It’s easy to see in all of my data sets.
One other thing that is important to note. All of my data are acquired on a plane. We see the flow only on the plane cutting through the center of the ball. If the action is in that plane, it is pretty easy to interpret. If there is strong motion out of that plane, some creativity is required.
1) If the wake of a ball is not symmetric, there is an aerodynamic force on the ball.
This is true no matter the source of the aerodynamic force, whether it be Magnus force or the new force I am describing here. Figure 1 shows PIV data of a 90 mph 3000 RPM fastball from the side. The seams are playing no role here. The deflection of the wake is purely from Magnus force. This wake is deflecting down, meaning that the ball is being pushed upward, against gravity.
Magnus force defects the wake more when there are more RPM, therefore making a bigger force. For comparison, the data in Figure 2 is for a 90 mph, 1650 RPM fastball. Note it looks similar, but the wake is not deflected as much.
2) Seams on the back of the ball can cause the flow to separate early or late, and this can lead to a shifted wake.
I have lots of examples of this. The one shown in Figure 3 is a knuckleball that is doing what it was born to do. The seam at the lower rear of the ball has pinned the boundary layer separation point. On the top, smooth side of the ball, the separation happens earlier, and the wake is shifted as a result.
Looking at hundreds of data sets, it is clear that there is a range of locations beyond the center of the ball where a seam will pin the boundary layer separation point. I’ve illustrated this schematically for a non-spinning ball (or one spinning our of plane) in Figure 4.
For a ball that is not spinning, the red regions are the same on the near and far side of the ball also. The size of the red region over which a seam may cause detachment is under investigation, but…
3) Rotation of the ball shifts the range of angles over which the flow can separate off a seam
These videos of 2- and 4-seam fastballs in Figures 5 and 6 may make it the most obvious. I’m plotting vorticity here (don’t worry about what it is, but if there is red or blue, the flow has separated and you are in the wake). The blue dots mark the locations where the boundary layer has separated. Full disclosure: each frame of these videos is a different ball. Nazmus stitched these together to give us a sense of how the flow looks over a full rotation of the ball. Note how, on the back of the ball, the flow separation is usually on a seam, unless the seam is too far beyond the center (I call it the hemisphere line, see Figure 4) of the ball. The separation points on the top seams are at a much larger angle than the ones on the bottom seam.
If we look at a fastball from the side, the range of angles over which the flow may separate on the top seam grow larger while the range of angles for the bottom seam become smaller. However, a fastball viewed from above should behave similar to a ball that is not rotating at all. Based on these observations, I made the sketch in Figure 7. Both red zones are shifted in the rotation direction.
4) The force indicated by a deflected wake may be temporary
Magnus force is quite steady. But, forces due to boundary layer separation at a seam on the back of the ball may not last long. If the seam leaves the area due to rotation of the ball, the force may disappear. Looking at Figures 5 and 6 the wake is quite often momentarily shifted more or less, but the ball will respond to the average deflection. Our PIV data are instantaneous, and they don’t tell us much about what happened just before or after we took the shot. If the force on the ball is changing with the ball’s rotation, for rotation rates in the thousands of rpms, there is no chance for the ball to change its direction due to the change in force.
5) The goal of this pitch is to park a seam on the red line (Figure 4) on one side of the ball but not the other, and to keep it there.
We have all seen a straight 2-seamer and noted how they appear to have circles on the sides. This ball orientation has the feature that the a seam appears near the red areas from Figure 4 most of the time. But, for a conventional 2-seamer, the effect is symmetric, so it generates no force. The circles may also fail to be in the “Red zone”. An example is shown in Figure 8. The seams are having no effect since they are both too far back. Even if the flow detached from them, it would be the same on both sides.
Here is another 2-seam fastball viewed from above. This ball came out of our machine a bit crooked, but the axis of the ball remains straight horizontal (indicated in Figure 9).
This wake is deflected down due to the presence of a seam near the lower red zone from Figure 4. However, note that this ball is spinning, and due to its axis, after 180 degrees, the upper seam will be in a red zone and the lower one will not be. In a case like this, the wake is rapidly oscillating and has little effect. (Side note, I think I can hear these pitches buzz–we’ll see).
Take a look at Figure 10 which is a still from the movie in Figure 11 (this was part of our Driveline study) and note the seam indicated by the red arrow. It is near the red region from Figure 4 (it may not be in the red zone, but that’s not the point right now). The gyro component of this pitch allows that seam to be located consistently in the same spot on the back of the ball. There is also a seam on the right edge of the ball. This likely is in the red zone and may deform the wake.
People keep asking me how to throw this pitch, which is awkward, since I have no idea. Hopefully the video helps with that question.
So, in my opinion, this is about how much gyro, and what sign of gyro, the pitch has. It is easy to see from the sketches in Figure 12 and Figure 4 that the amount of gyro, or how tilted the axis is when viewed from above, will strongly impact whether a seam lies in the red zone.
6) What to call it?
If this explanation is correct, it is incorrect to call the pitch a “laminar express.” For one, adding gyro would make the pitch less likely to be laminar, not more. But I do not believe that the flow over a baseball above 80 mph is ever laminar beyond the hemisphere line and have attempted to provide a different explanation of why a pitch thrown 2-seam with some gyro can behave funny. I don’t think I have proven it works this way as yet and do not care to argue with Mr. Bauer until I do. And, it’s important to note that people are throwing this pitch. I am only trying to change our understanding of how it works.
7) Future work
We currently have data for raised seam baseballs. We will determine the limits of seam influence on separation (i.e. the red zones) for these balls for fastballs from above (no Magnus influence), from the side (which has shifted zones, like Figure 7) and for knuckleballs, which are slower and have no spin.
Next, clearly this all needs to be repeated with MLB and/or NCAA baseballs.
8) Epilogue
I have really enjoyed feedback from readers of these posts and have learned a lot from them. If you have a question/comment, please make it.
