Comparison of Baseball Catching Models



This program allows you to compare the optical velocity control (OVC) model to the optical acceleration cancellation (OAC) model of baseball catching. The OVC model is described in Controlled Variables: Behavior as the Center Fielder Views It by Richard Marken. The OAC model is described in many sources, such as recent JEP:HPP comment by McLeod, Reed and Dienes.

The OAC model is currently the most popular model of baseball catching. This demonstration shows that the OAC model, though popular, is wrong. The model misses many balls that are easily caught by the OVC model. The problem with the OAC model becomes clear from watching the center graph in the display above. The center graph shows optical acceleration, Acc (large squares), and optical velocity, Vel (small squares) varying over time.

Note that Vel and, as a side effect, Acc, are kept nearly constant throughout the entire catch by the OVC (Velocity) control model. This is indicated by the fact that the plots of both Vel and Acc over time are nearly flat throughout the period of the catch. Note, however, the Vel is not kept constant throughout the entire catch by the OAC (Acceleration) control model. Acceleration is kept at zero throughout most of the catch, indicating that the model is working; it is successfully controlling acceleration. Unfortunately, controlling acceleration does not always result in a catch, indicated by the fact that the ball (the large open square in the left-hand graph) does not end near the fielder (the closed oval). This is because controlling acceleration does not always result in controlling velocity at a constant value that is the "right" (reference) value that will result in a catch. In fact, the OAC model is often keeping optical acceleration near zero while optical velocity is actually increasing at the beginning and end of the catch. In order to catch a ball, the fielder must keep the optical velocity of the ball constant near the reference velocity (the flat line Acc/Vel plot). The OAC model fails because it can successfully keep optical acceleration near zero while failing to keep optical velocity at the appropriate value.

In order to successfully catch a ball it is necessary to control optical velocity at a particular, fixed (reference) value throughout the period of the catch. The OVC model actively keeps optical velocity at this reference value and, in the process, also keeps optical acceleration at zero. The OAC model keeps optical acceleration at zero. A side effect of this control is that optical velocity is kept relatively constant, but often at the wrong value, or constantly increasing, resulting in a missed catch.