The “Knuckleball is thrown off the tips of the fingernails in a manner that inhibits spin (or at least when thrown with great skill). As shown in our post on 2-seamers and on one on the seam’s affect on boundary layer, the location of boundary layer separation tends toward the seams, which are moving around randomly during this pitch’s flight to the plate. This sometimes results in a completely crazy trajectory, such as in the video below (from Alan Nathan’s site).
http://baseball.physics.illinois.edu/KBall/Dicke
Our machine can pitch knuckleballs, and we acquired PIV data on quite a few of them. All of these pitches are 65 mph. These balls arrive in a random orientation, so what follows is a series of data sets in no order. I will comment above each one on notable features. All of these data sets were acquired in a vertical plane.
The boundary layer separates at the seam on the bottom of this ball, but not on the top. My guess is that the laminar boundary layer becomes turbulent due to the seam on top and remains attached. This ball is being forced downward.
The boundary layer separations for this ball are laminar top and bottom. The blue lower boundary layer even has a nice tight K-H vortex visible, This ball has nothing but drag.
This one looks to have a wake that is deflected either towards us or away and probably has a sideways force.