Although a change in running direction is one of the most important skills for ball game players, there is little information on the techniques involved, and the criteria used for technical evaluation. The purposes of this study were to develop indices for evaluating the techniques involved in changing the direction of running (i.e. cutting motion), and to compare the motions used by skilled and less-skilled groups to gain insight into the cutting motion technique. Twenty male university players of several ball games participated as subjects. They performed three kinds of 30-m zigzag runs with directional changes in angle of 30, 60 and 90 degrees, using the side-step and cross-step techniques. Their motions were videotaped with two digital video cameras operating at 60 Hz for three-dimensional motion analysis. The running time was measured by an electric timing device with photocell sensors. The velocity of the center of gravity (CG), angle of direction change, and angles of shank inward lean and thigh were calculated. We developed a cutting motion index that was calculated from the distance between the CG positions before and after cutting, the horizontal velocity of the CG at the beginning of cutting, and the elapsed time during cutting. The index gives a ratio of the time estimated from the horizontal velocity of the CG at the beginning of cutting relative to the actual elapsed time. The major results are summarized as follows: 1) For zigzag runs except for a 60-degree change in direction in the cross-step technique, there was no significant relationship between the cutting motion index and the angle of direction change. 2) In the side-step technique, the acceleration of the CG, the extension of the knee and hip joints, and the maximal forward lean angular velocity of the thigh in the second support phase were larger in the skilled group than in the less-skilled group. 3) In the cross-step technique, the deceleration of the CG and the flexion of the knee joint were smaller, but the acceleration of the CG and the maximal forward lean angular velocity of the thigh in the second support phase were larger in the skilled group than in the less-skilled group. These results indicate that the technique used to maintain running speed can be evaluated by the cutting motion index, and that the angle of direction change can be used as an index of direction change for the cutting motion. In the side- and cross-step techniques, a fast forward lean of the support thigh in the second support phase is likely to be an effective motion for maintaining the speed of the CG. In the side-step technique, extension of the knee joint in the support phase increases the forward lean angular velocity of the thigh. A fast forward lean of the support leg is effective in the cross-step technique.