Abstract: Rotation being the secret to success in table tennis, the trajectory prediction research hence should take the rotation into consideration. This study designed experiment to obtain the precise initial position coordinates, initial velocity and direction, rotating speed and direction. These nine initial data were taken as the input information of neural network and the precise placement coordinates as the output information. The correlation between input and output information is discussed by using artificial neural network algorithm thereafter. The experimental results show that the initial speed, rotation and coordinate with the ball placement coordinates are correlated.