Most of motions used in physical education and sports require generation of a great amount of mechanical energy and its effective use to achieve good performance. The purpose of this study was to investigate effects of technical training on the takeoff motion and mechanical energy-related parameters in the standing long jump. Twenty-one male adults were participated in the technical training session for two days, which was designed to improve the jumping technique and jumping distance of the standing long jump. Jumping motion of the subjects was videotaped with a high-speed video camera and ground reaction forces were measured with a force platform in pre and post training sessions. A two-dimensional motion analysis technique was used to calculate total mechanical work (Wtotal), effectiveness index of mechanical energy utilization (EI), joint force power (JFP) and segment torque power (STP).

After the technical training, the subjects increased the forward lean of the body at the instant of the takeoff. This change resulted in the more forward extended takeoff position and the increase in the horizontal takeoff velocity as well as the jumping distance. There were two types of change in mechanical energy-related parameters by the technical training. One was an EI+ type which increased EI but decreased Wtotal, and another was a Wtotal+ type which increased Wtotal but decreased EI. The EI+ type decreased the mechanical energy flow, especially to the trunk from the legs and the trunk backward rotation, while the Wtotal+ type increased both. These results indicated that the two-day technical training of the standing long jump was able to improve not only the jumping distance but also the takeoff technique, and that the effects could be categorized by the mechanical energy flow to the trunk from the legs and the change in the trunk rotation during the takeoff phase.

This study was intended to clarify the relationship between features of change velocity and performance of elite athletes in men's 400-m hurdles (400 mH). Races of 400 mH were videotaped with several video cameras panning from start to goal at 60 fields/s. After a time indication was superimposed on each field of the VTR images, the time at touchdown immediately after hurdling was recorded. Using the flash of the starter's gun recorded on the VTR image, each hurdle touchdown time from the start and each section (hurdle to hurdle) time was obtained. Each section was defined as follows: S1 was from the start to 2^nd hurdle; S2 was from the 2^nd to 5^th hurdle, S3 was the 5^th to 8^th hurdle, and S4 was from the 8^th hurdle to finish. Results of the regression analysis showed significant correlations between the 400 mH race time and the total number of steps from hurdle to hurdle (TNS; r=0.397, p<0.05), S2 time (0.673, p<0.001), and S3 time (r=0.865, p<0.001), and the rate of deceleration from S1 to S2 (r=0.367, p<0.05), and from S2 to S3 (r=0.421, p<0.01), and the rate of S3 time in the race time (r=0.463, p<0.01). These results indicate that it is important to decrease TNS and to prevent decreasing velocity in S2 and S3, to attain a high performance.

The kip maneuver on the horizontal bar is one of the basic skills of apparatus exercise, but it is said to be difficult for beginners to learn and execute. Although knowledge of kinetics is necessary to effectively teach the kip maneuver to beginners, there has been little research on kinetics, especially joint torques, during the kip maneuver. The purpose of this study was to investigate and compare the shoulder and hip joint torques of skilled and unskilled subjects performing the kip maneuver on the horizontal bar. The skilled subjects were made up of ten male varsity club gymnasts, while five male subjects who had no experience performing the kip maneuver comprised the unskilled group. All kip maneuvers performed by the subjects were videotaped and analyzed with two-dimensional videography to calculate trajectories of the center of mass, joint angles, angular velocities, torques, and work at the shoulder and hip joints.

Compared to the skilled group, the center of mass of the unskilled group was raised higher and closer to the bar at the instant of maximal forward swing. In addition, peak shoulder extension velocity of the unskilled group was significantly smaller than that of the skilled group (p < 0.05). Although shoulder extension torque of the skilled group increased during the backward swing phase, that of the unskilled group increased and was larger during the forward and upward swing phases than it was for the skilled subjects. The work done by the shoulder joint torque of the unskilled group was significantly smaller than that of the skilled group (p < 0.05), while there were no differences in work done by the hip joint and total work between the two groups. These results indicate that the difference between the skilled and the unskilled groups was not in the magnitude, but rather in the timing of joint torque exertion.

To improve the kip maneuver for unskilled performers, it is suggested that the shoulder extension and hip flexion torques should be exerted later and that exertion of shoulder extension torque should be emphasized.

The effect of precise adjustment of muscle coordination on jump achievements and kinematics in human vertical squat jump (SQJ) was evaluated quantitatively using a musculoskeletal model and forward dynamics approach. In order to evaluate the effect of altered muscle coordination, an optimal solution muscle stimulation time was first found through an optimization method. The optimal stimulation time of muscles was systematically altered by the interval of 0.1 ms in the range±50 ms. It was found that jump performance critically depends on precise muscle control, especially in terms of inter-muscular coordination. The sensitivity of fine tuning the control of individual muscles was found to be correlated with maximum isometric muscle strength. The minimal time interval that significantly disrupted a SQJ performance amounted to less than 1 ms. It was found that intermuscular coordination is crucial in maximizing jump performance. Mutual coordination of mm. vasti and hamstrings ; m. gastrocnemius and m. rectus femoris ; m. adductor magnus and hamstrings ; m. gluteus maximus, m. adductor magnus and m. rectus femoris was found to be of great importance for jump achievement. The role of coordination of specific muscles in SQJ and possible implications of our findings to physical training have been discussed.

The purpose of this study was to investigate three-dimensional kinetic changes in the support leg during the curve phase in long distance speed skating. Eight male long distance speed skaters performed 4000 m skating at the maximal effort, using an instrumented sensor klapskate. The skaters were videotaped with two synchronized cameras (60 fields/s) in the mid portion of the curve track by using a panning DLT technique. Push-off forces and three-dimensional coordinates of the segment endpoints and blades during the left stroke were collected over 4000 m distance to calculate kinetics of the left support leg in the first and second halves, the points of 650 m and 3450 m, by an inverse dynamics. The contribution of the knee joint to total mechanical work done by the leg joints decreased in the second half, and the positive work done by knee joint torque in the second half was smaller than that of the first half. In the first half, the hip and knee joints exerted large extension torques during the gliding phase, and the knee joint exerted large extension torque and power during the push-off phase. The knee extension torque after approximately 30% stroke and extension power of the knee joint during the push-off phase in the first half were larger than those of the second half. These results indicate that minimizing the decrease in the knee extension torque of the left support leg under fatigue condition is an important factor to maintain the skating speed in the curve phase of the long distance speed skating.

The purpose of this study was to investigate kinematic factors to pitch fast ball by comparing motions of two groups of baseball pitchers whose ball velocities were significantly different. Twenty-two baseball pitchers participated in this study as subjects. Their pitching motion was videotaped with two high-speed video cameras to compare three dimensional kinematics. Subjects were divided into two groups, i.e., high and low ball velocity groups (HG and LG). Average ball velocity of the HG was significantly larger than that of the LG (HG, 35.7±1.0 m/s, n=10; LG, 33.2±1.1 m/s, n=12; p<0.001). From the instant of the maximal knee height of the stride knee to the middle of the second phase (from the instant of the minimum height of the throwing hand to the instant of the stride foot contact), the HG flexed the knee and hip of the pivot leg and inclined the thigh backward more than the LG. At the stride foot contact (SFC), the HG flexed the knee of the stride leg and rotated the lower trunk more than the LG. In the first half of the third phase (from SFC to the ball release), the elbow joint angle of the throwing arm of the HG was kept about 80 degrees while it was about 60 degrees for LG. The timing of the peak angular velocity of the horizontal adduction of the shoulder of the HG was significantly later than that of LG, but the angular velocity was larger in the magnitude because of longer range of motion of the horizontal adduction in HG.

A lot of evaluation parameters of the static posturography for the age-related change of the postural stability has been presented to date. However, the time-dependent stochastic property and the dynamics of postural sway have not been adequately considered. The purpose of this study was to investigate evaluation parameters of postural stability change in the aging population considering the stationarity of variables. The participants were healthy adults (20-83 years, n=395). First, the stationarity of variables (the center of pressure (COP), the center of mass aceleration (ACC), and the differential series of COP (dCOP)) and the reproducibility of summary statistics of these variables were examined using data of 80 male participants below 60 years old. Then, the age-related change of these evaluation parameters was examined using all participants' data. The results indicated that, 1) ACC and dCOP were more stationary than COP; 2) the summary statistics of ACC and dCOP were highly reproducible; 3) the Hurst's exponent in the short time region (H_s) was highly reproducible, whereas the Hurst's exponent in the long time region (H_l) was not; 4) the mean velocity (V), which was equivalent to the mean amplitude of dCOP, and the squared mean velocity (V_s) of COP clearly increased accompanied with age; 5) there were also increasing trends with age for statistics of ACC; 6) H_l had a decreasing trend with age. We conclude that V, V_s and statistics of ACC are adequate as evaluation parameters to detect age-related change of postural stability, while the summary statistics of COP are not adequate as evaluation parameters because of un-stationarity of COP. Further, while Hurst's exponents have potential as evaluation parameters, it is probably difficult to obtain accurate values in the standard measurement protocol of posturography.

In an experimental study, the airflow conditions around a ski-jumper during a take-off motion was analyzed. The airflow conditions on the back of a jumper during the take-off were observed by the surface tuft method in a wind tunnel and in actual ski-jumping. The influence of the angle of attack (AOA: the trunk angle measured from a moving direction) during postural transition was investigated.

In wind tunnel test with the wind speed of 16 m/s, the jumper performed the simulated take-off with the low- and the high-AOA. The airflow condition around the vertebra prominence was always turbulent throughout the motion. With the high-AOA (approximately 30 degree), the airflow along the body-sides turned outwards to the body-sides, and the separation of airflow toward both sides was observed at the hip. On the other hand, these changes of airflow condition were not observed in the case with the low-AOA.

The aerodynamic lift and drag forces were calculated from the vertical and the horizontal ground reaction forces in the wind tunnel. In motions with the high-AOA, the aerodynamic forces, especially the drag force was larger (p<0.05) than the cases with the low-AOA.

The airflow conditions were measured in actual ski-jumping. 5 jumpers performed 18 trials at the Miyanomori ski-jumping hill (K=90 m). While a jumper passed a hill-edge, the turbulence of the airflow on the posterior side was generally small. After passing the hill-edge, the airflow condition changed apparently. As the jumper changed his posture, the similar airflow condition to that in the wind tunnel test was observed. The turbulence observed in the actual ski-jumping trials was larger than in the wind tunnel test.

Through this analysis, it was verified that the effect of the AOA on the airflow condition around a jumper is significantly large. The increase of the AOA causes turbulence and the airflow separation on the posterior side of a jumper, and as a consequence, the aerodynamic drag force increases. This suggests that the low AOA during a take-off is a prerequisite to attain a good ski-jump flight.

The purpose of this study was to analyze the mechanical energy during baseball pitching for 22 varsity baseball players by using three-dimensional motion analysis technique with two force platforms. Joint torque powers, joint force powers, and segment torque powers of the joints were computed using an inverse dynamics approach.

In the energy increasing phase of upper torso, a great deal of mechanical energy flowed into the torso. The mechanical energy transferred to the upper torso due to the segment torque power significantly related to the ball velocity at the release (r=0.480, p<0.05). In the late cocking and accelerating phase, great mechanical energy flowed into the distal segment and the ball due to the joint force power were observed at the throwing arm joints. There were significant relationships between the ball velocity at the release and mechanical energy flows due to the elbow and the wrist joint force powers (r=0.775, p<0.001 and r=0.827, p<0.001). These results suggested that the mechanical energy flows to the upper torso in the energy increasing phase of upper torso and to the throwing arm and ball in the late cocking phase are important to increase the ball release velocity.

In previous drop landing studies, support of the lower extremity was investigated by independent examination of the ankle, knee, and hip joint moments. Support of the lower extremity based on the net effect of the three joint moments has not been previously investigated. In addition, in those studies the horizontal ground reaction force was not examined. On the other hand, Winter (1980) proposed a support moment calculated from the summation of the ankle, knee and hip joint moments in a walking study. The purpose of the present study was to investigate correlation between the support moment and the lower extremity joint moments, as well as the effects of the joint moments on the horizontal ground reaction force, via sensitivity analysis.

Eight healthy male subjects (age; 25.30±2.71 yr, body mass; 67.52±6.25 kg, and height; 175.60±3.07 cm) participated. A force platform (Kistler type 9281 B, Kistler Instruments, Switzerland) was used to measure ground reaction forces (GRF) at 1 kHz. An imaging system (3D OPTTRACK, Northern Digital Inc, Canada) was positioned 5 m from the force platform at 0.2 kHz. Each subject performed drop landings from two platforms; 0.38 m (LOW) and 0.61 m (HIGH) from the floor. Subjects placed their hands on their hips while stretching out one leg above a target landing position on the force platform, and pushed off from the box with their leg on the box. Then, the legs were closed in the air, subjects landed on the force platform with both feet simultaneously. Joint moments were calculated using an inverse dynamic analysis. In addition, the sensitivity of the joints moments and the support moment to changes in the horizontal GRF were analyzed.

Significant positive correlation was observed between the hip joint moment and the support moment at the maximum peak of the support moment (r=0.838, p<0.001). The support moment also showed significant positive correlation with the vertical GRF (r=0.865, p<0.001) and the horizontal GRF (r=0.699, p<0.001). In the sensitivity analysis, when the horizontal GRF was changed, the hip joint moment changed more markedly than the others. Thus, these findings indicate that decreasing the horizontal GRF is an important factor in decreasing the support moment.

The purpose of this study was to confirm quantitatively the effect of foot orthotics on the reduction of ground reaction forces during running. Twenty-seven volunteers participated in this study. Ground reaction force (GRF) data were obtained from the subjects while running in two insole conditions (control and orthotic ones). The foot orthotics used in this study had mechanical functions to support medial longitudinal arch and rearfoot. Peak sideward GRF around heel-contact significantly reduced at 32% for the orthotic condition compared to the control one (P < 0.05). Cluster analysis based on the time series of sideward GRF data enabled to categorize the subjects into four representative groups. Peak sideward force significantly reduced at 54% for the orthotic condition (P < 0.01) for the group in which the subjects showed high heel-contact impact (more than 2.5 N/kg). These results suggested that well-designed foot orthotics could effectively reduce sideward impact force around heel-contact during running.

The purpose of this study was to estimate the knee extension torque generated by vastus lateralis (VL) during pedaling on the basis of VL tendon elongation (ΔLt). The fascicle length and pennation angle of VL during isometric knee extensions (0, 20, 60, 100 Nm; 30°, 45°, 60°, 75°, 90°, 105°) were measured using a B-mode ultrasonography (n = 5). ΔLt was estimated from these measurements and the relations between ΔLt, knee joint angle and the knee extension torque generated by VL (TQ) were obtained. ΔLt of the same subjects during pedaling was taken from a previous study. TQ during pedaling was estimated on the basis of ΔLt during isometric knee extensions and pedaling (TQ_ΔLt). TQ during pedaling was also estimated on the basis of the inverse dynamics (TQ_inv). The peak value of TQ_ΔLt (19.8 ± 11.2 Nm) during pedaling was significantly greater than that of TQ_inv (9.3 ± 4.0 Nm) (p < 0.05), which might be related to the lower activation level and the different activation pattern of the rectus femoris compared with those of the VL. The pattern of TQ_inv differed significantly from that of TQ_ΔLt, which might be related to the activation of the biceps femoris that increased in the latter half of the knee extension phase during pedaling. Since the methodology used in this study is non-invasive and measures the parameter that is closely correlated to the joint torque generated by a single muscle, it can be a useful tool for investigating the mechanics of human movements.

In the present study, muscle fiber behavior of the medial gastrocnemius (MG) and soleus (SOL) muscles during isokinetic concentric (CON) and eccentric (ECC) plantar flexions were directly measured in vivo. The subjects (n = 6 men) performed maximal voluntary plantar flexions on an isokinetic dynamometer at 60, 120, 180, 240 deg/s, concentrically and eccentrically. With the use of ultrasonography, muscle fiber lengths and pennation angles were determined. From the time course of fiber length changes, muscle fiber shortening and lengthening velocities (V_fiber) were obtained. Considering the effects of pennation angle, muscle shortening and lengthening velocities were also calculated. V_fiber changed as a function of angular velocities [MG: 32-90 mm/s (CON) and 37-97 mm/s (ECC); SOL: 35-81 mm/s (CON) and 37-84 mm/s (ECC)]. However, relative changes of V_fiber were less than those of angular velocities. This result suggests that muscle fiber behavior does not always match joint movement due to muscle-tendon interaction. V_fiber of MG was larger than that of SOL for all test conditions, suggesting inter-muscle differences in muscle fiber behavior during isokinetic movements, but the differences were not evident at the muscle level.

The purpose of this study was to investigate the change in kinetics of the lower limb joints in 100 m sprint running. Nine male sprinters, running 100 m with the maximum effort, were videotaped at every 10 m mark from the start to 90 m mark with five high-speed cameras (250 Hz) and five normal VTR cameras (60 Hz). Two dimensional coordinates of the body landmarks were obtained by digitizing VTR images over at least, two steps at every 10 m. Performance descriptors, kinematics and kinetics such as stride length, stride frequency, joint torque and joint torque power of the lower limb joints were calculated by an inverse dynamics approach, with estimating ground reaction forces from the acceleration of the whole body center of gravity. The relationships between changes in running velocity and the joint torque and joint torque power of the leg were analyzed. (1) 100 m sprint could be divided into four phases: the first and second acceleration phases, the maximum speed phase, and the deceleration phase. (2) In the first acceleration phase, 0-30 m, where the stride length and stride frequency increased. There were significant relationships between increases in running speed, stride length, and stride frequency and increases in negative power exerted by knee extensors in the early recovery phase, positive power by hip extensors and negative power by knee flexors in the late recovery phase. (3) In the second acceleration phase, 30-50 m, where the stride length increased and the increase in the running speed was smaller than that of the previous phase. Positive power exerted by hip flexors in the early recovery phase and negative power by knee flexors decreased. (4) In the maximum speed phase, 50-70 m, significant increase in powers to the previous phase were found in positive power exerted by hip flexors in the early recovery phase, negative power by knee flexors in the late recovery phase, and power by plantar flexors during the support phase so as to allow the sprinters to acquire higher speed. (5) In the deceleration phase, 70-100 m, where the stride frequency as well as the running speed decreased. The decreases were found in positive power exerted by hip flexors in the early recovery phase and negative power by knee flexors in the late recovery phase.

These results revealed that the critical factors in the joint kinetics of 100 m sprint running would be power exertions of hip extensors, knee extensors and knee flexors in the late recovery phase for the acceleration, of hip flexors in the early recovery phase, and of knee flexors in the late recovery for high speed and speed endurances.

In this study, the electrical activity patterns of the antagonistic bi-articular and mono-articular muscles of upper extremity, during in the hand contact period of in front handsprings in tumbling, were analyzed in terms of electromyographic (EMG) kinesiology and robotics. The mechanical two-joint link models equipped with a pair of the antagonistic bi-articular muscles in addition to two pairs of the antagonistic mono-articular muscles were used for robotics analyses. Reversal of triceps brachii long head (Tlo) and biceps brachii long head (Blo) activities was responsible for changes in the output force direction during the hand contact period. It was confirmed that the coordinating activity pattern of three pairs of the antagonistic muscles contributed to the output force control and the output force direction control. The results obtained here suggest the importance and necessity for coordinated muscle activities of bi-articular and mono-articular antagonistic muscles for analyses of sports.

Purpose of this study was to estimate the moment arm of human Achilles tendon by application of gravity. Eight male subjects were requested to perform passively plantar-and dorsiflexion movements by a dynamometer with two different leg positions, i.e. one was the horizontal position of leg and thigh with the knee joint fully extended (setA), and the other was the supine horizontal position with 60° leg up (setB). Excursion of myotendinous junction (MTJ) was determined by ultrasonography during passive ankle movement. Moment arm was calculated by dividing the MTJ excursion (mm) by the angle change (rad). At the plantarflexed position (60°), the moment arm in the setB (35.1 ± 5.6 mm) was significantly higher (p < 0.05) than that in the setA (26.0 ± 10.0 mm). The moment arms obtained in the setB were within a range of reference data reported in previous studies, suggesting that the application of gravitation is useful to estimate Achilles tendon moment arm.

The effectiveness index of mechanical energy utilization (EI) is defined by the following equation;

EI =

　　Performance and / or Effective energy (and / or Work)

　　　　　　 Mechanical energy (and / or Work) 　　　　

The generic definition of EI is useful for evaluating skilled motion in human movement from the viewpoint of effective usage of mechanical energy and/or work.

Certain correlations between EI and locomotion speed were observed in several locomotion studies using EI. If the EI has correlations with locomotion speed, however, interpretation of the EI is complicated. In the present study, EI was applied to the evaluation of walking and running motion in order to examine if EI correlates with locomotion speed.

Two subjects were asked to perform 15 walks and runs at several speed levels. The trials in each speed level were measured using an automatic coordinate acquisition system and a force platform. Then the EIs with several numerators representing performance and/or effective energy utilization were calculated. The numerators were P_E (= 0.5 x body mass x (locomotion speed)²), P_V (= body mass x locomotion speed)_, P_D (= body mass x step length) and P_DV (= body mass x step length x locomotion speed). W_JP (absolute work in a cycle obtained based on joint torque power) was used as the denominators of EI.

The all EIs examined in this study have significant correlations with locomotion speed. In these cases, correlation coefficients as well as regression coefficients were different among subjects, movements and the numerators of EI. Based on these results, in order to appropriately examine EI of the target locomotion, firstly we need to confirm correlation between EI and the locomotion speed. Then, if EI correlates with locomotion speed, it may be necessary to consider the effect of locomotion speed in the interpretation of EI.

This study was conducted to investigate the kinematic characteristics of javelin throw for further comprehension of the techniques and mechanics of the javelin throw. Forty-nine male javelin throwers were videotaped, using two 200 fps high-speed video cameras in three athletic competitions.

The results revealed the significant positive correlation between the initial velocity of javelin and the throw distance (r = 0.797, p < 0.001). The throwers showing the higher initial velocity of javelin showed the higher forward rotation angular velocity of trunk at the moment of javelin release (r = 0.484, p < 0.001). The elbow joint angle and shoulder abduction angle at the moment of javelin release showed the significant negative correlations with the initial velocity of javelin (r = -0.437, p < 0.01 and r = -0.454, p < 0.01, respectively). The velocity calculated by taking the shoulder velocity from the grip velocity at the moment of javelin release showed the significant positive correlation with the initial velocity of javelin (r = 0.819, p < 0.001). These results suggested that both extension and internal rotation of the shoulder contributed to the enhanced relative arm velocity for the elite throwers, while the internal rotation of the shoulder for the novice throwers limitedly contributed to it.

The kip maneuver is one of the principal skills of apparatus exercise at the horizontal bar, but it is difficult for beginners to execute the maneuver. Most of the studies on the kip maneuver compared successful trials with unsuccessful ones, or the trials of skilled subjects with that of unskilled ones. Although skill training is mandatory for beginners to successfully complete the kip maneuver, there is no study on the process of learning and the effect of skill training on the changes in biomechanical parameters in the kip maneuver. The purpose of this study was to investigate effects of skill training for the kip maneuver on the shoulder and hip joint torques in the learning process. The kip maneuver of ten male varsity club gymnasts was videotaped and analyzed two-dimensionally to make the motion model of the kip maneuver of the skilled subjects. Three male subjects with no experience of performing the kip maneuver were selected as unskilled subjects. They participated in skill training for 5 to 12 days to learn the kip maneuver. All trials of the unskilled subjects were videotaped during the training. We selected every five trials to analyze changes in joint angles, torques of the shoulder and hip joints, and mechanical work done by the joints.

The number of trials the unskilled subjects needed to successfully complete the kip ranged from 47 to 108. Movements of the shoulder and hip joints in the trials in the second half of training were similar to those of the successful trials of all unskilled subjects. The range of the hip movement gradually increased with the number of trials. In the skilled subjects, immediately after the reversal point of the swing, i.e. the forward to the backward swing, the shoulder extension torque was increased rapidly with the extension of the shoulder joint, while the hip extension torque was exerted continuously in the same phase. This may indicate that the skilled subjects extended the shoulder joint strongly and intentionally. In the first trial of the unskilled subjects, the shoulder joint torque had three peaks. The hip extension torque was very small, whereas in the trials from the midpoint of the training, one of the peaks of the shoulder joint torque disappeared and hip extension torque was developed during the backward swing. In the unskilled subjects, the maximal hip extension torque significantly related to the number of trials in subjects B (y = 158.5 + 0.99 x, p < 0.05, where y is the maximal hip extension torque and x is the number of trials) and C (y = 145.3 + 1.09 x, p < 0.05, y and x respectively). The mechanical work done by the shoulder joint showed no significant relation to the number of trials in all subjects. Although the mechanical work done by the hip joint significantly related to the number of trials (subj. A: y = 150.2 + 2.15 x, p < 0.05; subj. B: y = 164.3 + 1.30 x, p < 0.05; subj. C: y = 196.7 + 1.22 x, p < 0.05, where y is the mechanical work and x is the number of trials). These results may suggest that it is essential for the unskilled subjects of the kip maneuver to increase the hip extension torque and the mechanical work done by the hip joint as well as to exert the shoulder extension torque.