An April 29, 2011, a low split-finger fastball thrown by Yankees right-handed pitcher Freddy Garcia inspired a whole science study.

As I have discussed in the past (I am plagiarizing myself from the linked article but I put it succinctly the first time), baseballs balls curve for a reason; the Magnus Effect shows us that the backspin of a fastball creates a high-pressure zone in the air ahead of and under the baseball due to the baseball's raised seams. The topspin of a curveball creates a high-pressure zone on top of the ball, which deflects the ball downward in flight, which combined with the awesome power of gravity, gives the ball more of a drop.

Garcia's split-finger fastball should have reduced backspin on the ball and prevented the ball from dropping much. The typical Magnus effect on the ball will tilt it slightly in toward the batter.

Magnus Force
Velocity           Drag

Spinning ball

But it didn't do that at all. Instead, it moved away from the batter.  Writes Zach Schonbrun in the New York Times, a writer at Baseball Prospectus saw it and was puzzled enough to reach out to a physicist for an answer, who contacted another physicist, Rod Cross, from the University of Sydney  - tests were done, a paper got written.
The key for Garcia’s pitch, Cross said, is enabling the ball’s spinning axis to pass through this smooth patch for as long as possible. Garcia manages to take enough spin off his splitter and spin it in a way different from that of any other pitcher.

What he theorized is termed the smooth patch effect. When air travels over a spinning baseball, its flow is disrupted by the raised and rough seams. That turbulence applies a force on the ball, causing it to break. Wherever the ball is smooth, however, or not covered by the seams, will cause it to go away from that direction.
Some curves in baseball are more illusion than physics.

Nice to see science make sense of the crazy things that happen in sports this time.

Citation: Rod Cross, 'Aerodynamics in the classroom and at the ball park', American Journal of Physics, April 2012, Volume 80, Issue 4, pp. 289 http://dx.doi.org/10.1119/1.3680609

Zach Schonbrun, New York Times

H/T David Dobbs