Twenty-four male university baseball players were each requested to throw a baseball, and filmed using the direct linear transformation method of three-dimensional (3D) videography. 3 D coordinates of landmarks were obtained. Resultant joint forces and resultant joint torques in the wrists, elbows, shoulders, neck, and upper torso joints were calculated using the inverse dynamics method. The mechanical powers caused by the resultant joint forces (joint force power) and by the resultant joint torques (joint torque power) of each segment were calculated, and the mechanical work was also obtained by integrating the joint torque powers with time. Peak values of energies of the upper torso, upper arm, forearm, hand, and ball appeared in sequence from the proximal segment to the distal segment. The joint force powers in any segment were markedly larger than the joint torque powers. Little joint torque power was produced in the wrist throughout the throwing motion. The negative joint force power and joint torque power at the proximal end of the upper torso were rapidly increased immediately after the foot contact stride. It was clarified that the appearance of the large energies in the distal throwing arm segments during the final phase of throwing motion were caused mainly by transfer of the energies produced by the motions of the torso and shoulder joints. This paper discusses the mechanical energy flows of the upper torso and upper limb segments during the motion of baseball throwing.

This study was designed to clarify the causes of throwing injuries of the elbow and shoulder joints in baseball. Five varsity-skilled baseball players without pain in the elbow and shoulder joints were subjects for this study. They were fixed to a chair and asked to throw a baseball using three different throwing arm movements (T₀, T₄₅, and T₉₀) . These movements were filmed using three-dimensional DLT videography. Linked rigid-body segment inverse dynamics were then employed to determine resultant joint force and torque at the elbow and shoulder joints. Peak varus torque at the elbow joint for T₉₀ was less than for the other movements during the acceleration phase. In the follow-through phase, however, a large anterior shear force (70 N) at the elbow, for elbow extension, was present for T₉₀. These results indicate that T₉₀ was a high risk movement which leads to extension injuries rather than medial tension injuries. After the ball release, a large superior shear force (118 N) at the shoulder joint was present in all movements. This superior force may result from the subacromial impingement syndrome, except for critical zones of impingement caused by the different throwing arm movements. These findings suggest that the mechanisms of throwing arm injuries are closely related to differences in throwing arm movements.