In professional baseball and major league baseball, pitchers often use a breaking ball called a forkball or split as a decisive pitch. These balls are known to be difficult to hit because the batter feels as if the ball is falling rapidly at the plate. Recently, measurement technology has advanced to the point where it is now possible to get detailed information on the speed, rotation and trajectory of the ball. While a backspin four-seam straight ball hops more than a parabolic trajectory due to the Magnus effect, a two-seam forkball is closer to a parabolic trajectory despite having the same backspin. Thus the mystery of “why does it fall more than the four-seam?”

A team led by Professor Takayuki Aoki of Tokyo Institute of Technology (Titech) has conducted research using the supercomputer TSUBAME3.0 at the Global Scientific Information and Computing Center, and performed a numerical fluid dynamics simulation of a baseball flying at high speed while rotating, calculating the seams in detail. A similar analysis was conducted 20 years ago by Dr. Ryutaro Himeno of RIKEN, who was the first to analyze a changing ball using supercomputing resources. The results of this analysis seem to confirm the results of Dr. Himeno’s analysis 20 years ago.

The aerodynamic analysis of a baseball requires the use of high-resolution computational grids, and the ability to efficiently place high-resolution grids near the surface and in the wake of the ball, as well as the ability to set accurate boundary conditions for moving seams, contributed greatly to the realization of this numerical simulation.

A computational fluid dynamics simulation using the supercomputer TSUBAME3.0 was performed for a low-speed rotating forkball with a two-seam backspin seam. As a result, the research team found for the first time that the ball does not approach a parabolic trajectory due to the low speed rotation and weak upward lift force, but that the trajectory is lowered due to the downward force “negative Magnus effect” occurring in the seam angle range of -30 to 90 degrees.  They also found that the “negative Magnus effect” does not occur with a four-seam pitch of the same velocity and rotation speed.

They also found that if they knew the velocity, rotation speed, and axis of rotation immediately after releasing the ball, the trajectory of the ball could be accurately reproduced. For a two-seam ball with a velocity of 150 km/h and a rotation speed of 1,100 rpm and a four-seam ball with a velocity of 150 km/h and a rotation speed of 1,100 rpm, the drop of the ball at the plate differed by as much as 19 cm just because the seam was different.

In addition to baseball applications, this analysis is expected to be used in tactics for non-rotating soccer balls and volleyballs, as well as in winter sports, where aerodynamic forces have a strong impact. In addition, this numerical simulation can be applied to various industrial fields.