The rotation of the shoulder and pelvis plays a significant role in induced seoi-nage performance. If the opponent cannot pull adequately or lacks strength, shoulder and pelvis rotation cannot occur sufficiently. In this study, we separately measured the horizontal rotation angles of the shoulder and pelvis to identify the kinetic characteristics of the torso rotation force. The horizontal rotation angle of the shoulder and pelvis of elite judo players was higher than that of amateur players. In other words, the unbalanced movement was sufficiently performed, thereby resulting in a flexible rotation of the shoulder and pelvis. Studies by Franchini [11] and colleagues demonstrated that elite judo players display higher body rotation in seoi-nage than amateur-judo players. Therefore, the better the body rotation, the higher is the influence of seoi-nage. However, in Event 2, the horizontal rotation angle of the shoulder was smaller in EG than in AG. It supposedly focused more on tilting upward and pulling toward the body than turning the shoulders. In the overall progress from Event 2 to Event 4, an analysis of the front and back tilt angle of the shoulder and pelvis revealed that the rear tilt was higher in EG than in AG. With seoi-nage progression, the front tilt was significant. Hence, the range from the rear to the front tilt was higher in EG than in AG. In other words, the shoulders of the EG supposedly tilted more backward for unbalancing and positioning than those of the AG and subsequently forward for an effective hang-up operation of seoi-nage.
In contrast, in Event 2, the shoulder displayed a smaller back tilt angle in AG than in EG. Moreover, with the skill progression, the front shoulder tilt was smaller in AG than in EG. Thus, insufficient back shoulder tilt for unbalancing and positioning was absent. Moreover, the front tilt of the shoulder for seoi-nage appeared small, such that differences could be observed according to the induced proficiency. The pelvis and shoulders tilt the same pattern, which is thought to show the difference in movement according to unbalancing and positioning. In other words, appropriate rear flexion of the shoulder and pelvis facilitated pulling the opponent toward the body. The front tilt of the shoulder and pelvis enabled carrying the pulled opponent forward [6, 12, 13, 14, 15]. The shoulder and pelvic tilt range may be related to the inertial force required to position the opponent. The moment of inertia can also vary depending on the induced proficiency and can be a significant kinetic indicator determining the level of seoi-nage. We did not directly measure the moment of inertia. However, the current study results were consistent with those of Ishii [17] and colleagues, who reported on a more significant range of tilt of the shoulder and pelvis from unbalancing to positioning, considering that the size of the shoulder and pelvic inflection range can be inferred. An analysis of the articulation angles of the hip and knee joints demonstrated that the angles were larger in EG than in AG in all events, except in Event 2 for the right hip joint. The hip joint was also higher in EG than in AG because the shoulder and pelvic posterior tilt was higher in EG than in AG in Events 2 and 3. However, in Event 4, which is a hang phase, the left hip joint was also more extended than the left hip joint, and the front tilt of the shoulder and pelvis was more significant in the EG than in the AG. To summarize, despite the slight curvature of the hip joint, the sizable forward tilt of the shoulder and pelvis occurred because the body was kept parallel to the opponent by pulling the opponent substantially close. Moreover, the heel was lifted to enable the body to lean forward and throw. Owing to the experimental equipment, the hip joint flexion after positioning was not comparable. This is because we measured only the initial stage of the positioning and not the completion of the positioning.
Ishii [17] and colleagues mentioned that a flexibly bent hip joint at the initial point of positioning did not display sufficient angular velocity while performing the actual throw, thereby making it difficult to effectively perform seoi-nage. In addition, early stages of positioning necessitate a certain degree of the extended hip joint angle using sufficient angular velocity [13, 16]. Therefore, the extended hip joint angle in elite judo players can lead to more effective throws than amateur players. In addition, there was no significant difference between the groups; however, the flexion of the right hip joint in Event 2 was higher in EG than in AG. This may be attributed to the close positioning of the right knee to the opponent while lowering his posture during unbalancing, such that the right knee joint could bend naturally.
There was no significant difference in the left knee joint angle between the groups. However, in Event 2, EG tended to be more curved in the initial stage of performing seoi-nage than AG. In Event 4, EG displayed a more advanced knee joint angle than AG. This is because knee joint flexion should appear from unbalancing to the throwing phase. While the posture must be lowered below the opponent's center of gravity, the knee joint extension must appear to lift the latter [9, 17, 18]. In addition, the flexion and extension angles of the right knee joint were more extended in EG than in AG in all events, which is thought to be because the EG lifted or pulled the opponent's center of gravity as much as it tilted and flexed. Moreover, in Event 4, the extension of the left and right knee joints of EG revealed a significant difference. The right knee extension occurred because of the significant rotation of the shoulder and pelvis owing to the nature of seoi-nage. Previous studies have mentioned that the lower right knee extension in the positioning phase while performing seoi-nage did not display sufficient rotation of the shoulders and pelvis [14, 19]. Therefore, EG supposedly performed seoi-nage by sufficiently rotating the shoulder and pelvis.
In addition, flexion angles of the left and right knee joints for both groups demonstrated that the knee joint of EG was more flexed or similar in Event 2 than in Event 3. However, AG was unbalanced. In other words, both knees must be bent at unbalancing and tend to be maintained or slightly extended. Therefore, it is thought that the AG did not perform effective unbalancing and then positioning when performing seoi-nage, which resulted in a minor shoulder and pelvic rotation compared with EG. The horizontal rotation angle and tilt angle of the shoulder in the kinematic characteristics at the point of maximum force development revealed a difference of 36° and 16.52° for EG and AG, respectively, resulting in higher rotation and greater forward tilt. The pelvis rotation and tilt angles were 38.2° and 10.78° for EG and AG, respectively, thus revealing a similar pattern as the shoulder kinematic variables. EG displayed a more extended pattern than AG at the hip and knee joint flexion angles, thus indicating it appeared identical to the extended pattern required in the positioning phase. The maximal force vector sum was 102.8 N, which was larger in EG than in AG, thereby suggesting that EG was superior to AG in terms of the magnitude of the pulling force. In the kinematic characteristics during the maximum force appearance, the shoulder's horizontal rotation and tilt angles displayed 36° and 16.52° differences, respectively, compared with EG, indicating that more significant rotation and greater forward tilt were observed. The pelvic rotation and tilt angles also revealed a difference of 38.2° and 10.78° in EG and AG, respectively, thus displaying a pattern similar to the kinematic variables of the shoulder. EG displayed more extension in the hip and knee joint flexion angles than AG, indicating it appeared in a similar pattern as the extension required in the throwing phase. These results suggested that EG was superior to AG in terms of the above-mentioned kinematic characteristics. In summary, the vector sum of forces was more significant in EG than in AG. In addition, EG displayed a greater rate of change in horizontal rotation and the tilt of the shoulder and pelvis than AG. Moreover, it demonstrated flexion and extension patterns of the hip and knee joints that were more effective in throwing.
However, there are limitations to this study. We did not examine the strength characteristics of the right and left arms separately. Therefore, future studies should study the strength characteristics of each arm, in addition to equipment, to generate more useful primary data. Future research should also analyze the force characteristics, kinematic characteristics, and correlations of various judo throwing techniques by using throwing vector equipment to provide more critical primary data in the learning and training methods of the judo throwing technique.