Wrestling and judo are similar in certain aspects-both involve grappling, are, respectively, contact and combat sports, and use the weight category system. In wrestlers and judokas, some researchers have published several articles on the measurement of their muscle strength under various conditions at various regions (e.g., trunk, arm, knee, neck, and shoulder) (7,8,13,18,23). Similar to the case of other sports, the importance of the trunk region, in particular, was indicated in both these sports. However, there is limited research on the trunk muscles of athletes, and few studies have compared the trunk muscles of wrestlers and judokas.
Regarding the trunk motions during wrestling, wrestlers are required to strengthen sagittal movements such as the flexion and extension motions in order to assume the low posture that is unique to wrestling and to tackle and lift an opponent during practice and competitions (10,11,20,25). On the other hand, for judokas, the trunk rotation and lateral flexion motions have been reported to be very important for the standing and groundwork techniques involved in judo (5,9,21). There is a strong possibility that the differences in the specific characteristics of the trunk muscles resulting from the differences in the motion of the trunk region that are crucial to the 2 similar competitive sports can affect the trunk muscle cross-sectional area (CSA) and trunk muscle strength of wrestlers and judokas. It is very important to note that information on the anatomy and function of the trunk muscles is available to these athletes and their strength and/or technical training coaches to enable them to understand the sport-specific trunk muscle characteristics and to allow them to apply the concepts that strengthen the trunk muscles involved in each sport. Further, such an evaluation of the trunk muscles is extremely essential from the viewpoint of preventing sports injuries because low back pain has been a frequent complaint among active athletes as well as among wrestlers and judokas (10,15).
Athletic performance is integrated by many factors that include trainable factors (physiology, psychology, and skill), teachable factors (tactics), and other factors outside the control of the athlete and coach (genetics and age) (19). Of course, dependent factors differ from athletic events in varying degrees. Since wrestling and judo are individual contact and combat sports, respectively, as stated above, athletic performance in these sports is closely influenced by physical factors such as muscle strength, power, and skill. It is interesting to note whether trunk muscles can affect athletic performance. However, there is a lack of scientific evidence that estimates the effect of trunk muscles on athletic performance levels of the sport as well as in athletic events.
Magnetic resonance imaging (MRI) and computed tomography (CT) are the main methods used to analyze the CSA of trunk muscles (17,22,24). The absolute evaluation of skeletal muscle tissue can be obtained from the images of the CSAs of the muscles. Different methods and types of equipment have been used to measure the muscle strength of athletes (4,16,26). Although multiple joints and position setups render an appropriate evaluation difficult, trunk muscle strength has also been studied using special equipment such as Cybex or Biodex (1,4,7,10). Because of such methodological difficulties, a few studies exist on the evaluation of the trunk muscle flexor and extensor strength among athletes.
Athletes gain greater trunk muscle strength and their muscles have larger CSAs as a result of continuous daily training. This tendency can be considered a sport-specific phenomenon, i.e., different muscular characteristics should exist between wrestlers and judokas. As far as we know, there has not been a study that examined sport specificity on the trunk region, especially the CSAs of trunk muscles. In this study, it was hypothesized that in wrestling and judo, the CSAs of trunk muscles and trunk muscle strength are associated with the sport-specific characteristics in each sport. Moreover, to clarify relationships between athletic performance and the CSAs and muscular strength of trunk muscles, we also compared the CSAs of trunk muscles and trunk muscle strength with athletic performance levels in wrestling and judo.
Experimental Approach to the Problem
To prove our hypothesis, physical characteristics, the CSAs of trunk muscles, and various isokinetic trunk muscle strength parameters were measured using various devices. First, we compared all the parameters to confirm the sport specificity of the CSAs of trunk muscles and trunk muscle strength in wrestlers and judokas. Furthermore, this study also determined whether the CSAs of trunk muscles and trunk muscle strength were associated with athletic performance in wrestling and judo. Therefore, all the parameters were compared on athletic performance levels in each sport.
Twenty-eight elite collegiate male athletes volunteered to participate in this study (mean ± SD: age, 19.7 ± 1.2 years; height, 169.4 ± 4.5 cm; weight, 68.9 ± 4.8 kg; sport history, 8.5 ± 4.9 years). These participants comprised 14 collegiate wrestlers and 14 collegiate judokas, none of whom had experienced any lower back problems at least 6 months prior to the study. The 2 groups of subjects were determined to be of similar body weight. All the subjects regularly spent a total of approximately 3 hours practicing the sport; usually, they had 2 training sessions per day 6 days per week.
The present study protocol was approved by the Ethics Committee of the Nippon Sport Science University. Signed informed consent was obtained from all subjects prior to their participation. Further, the subjects were briefed about the objective of this study and its potential risks.
Anthropometric data of the subjects was recorded (height to the nearest 0.1 cm and weight to the nearest 0.1 kg). The age of the subjects and their sport history were also recorded. Sport history is defined as the period of time that a subject practiced the sport before participating in the present study.
Cross-Sectional Areas of Trunk Muscles
To evaluate the CSAs of trunk muscles, MRI was performed using a 0.3-T magnetic resonance system that uses surface coils with a body coil (AIRIS; Hitachi Medical Corp., Tokyo, Japan). The subjects lay on a bed in the MR imager in a comfortable relaxed supine position. Transverse MR spin-echo T1-weighted images were obtained at the L3-4 level parallel to the lumbar disc space (TR [repetition time], 760 ms; TE [echo time], 20 ms; matrix, 256 × 265; field of view, 320 mm; slice thickness, 5.0 mm).
In order to measure the CSAs, the image was traced onto a paper, as shown in Figure 1; the traced image was transferred to a computer. The CSAs were calculated using image analysis software (Scion Image Beta 4.02; Scion Corp., Frederick, MD). In this study, we grouped the CSAs of the trunk muscles into 5 areas because they had poorly defined borderlines. As shown in Figure 1, each of the 5 areas comprised the same CSA on the left and right sides (rectus abdominis [RA]; oblique muscles [OB]; psoas [PS]; quadratus lumborum [QL]; paraspinal muscles [PA]). Two of these areas included multiple muscles; OB included the internal and external obliques and the transversus abdominis muscles and PA included the erector spinae and the multifidus muscles (3,6,24). Both the absolute and relative values were included in the CSA parameters and were normalized by dividing them by the subject's body weight.
Trunk Muscle Strength
The isokinetic of trunk flexor and extensor strength with peak torque, work, average torque, and average power were measured using a Biodex System3 with a back attachment (Biodex Corp., Shirley, NY). For the purpose of measurement, the subjects assumed a semistanding posture with their knees flexed at 15°. The axis of movement in the extension-flexion cycle was set at the L5-S1 level. The subject was secured to the device by an attached chest harness, pelvic support, and a thigh strap. The procedure was explained, and the subjects performed the standardized warm-up. Then, the subjects performed 3 reciprocal cycles of trunk extension and flexion at angular velocities of 60°·s−1, 90°·s−1, and 120°·s−1 with a 100° range of motion. The best trial value that was obtained from the 3 cycles was used. The trunk muscle strength parameters as well as the CSA parameters were evaluated as the absolute and relative values.
To analyze the CSAs of trunk muscles and the influence of trunk muscle strength on athletic performance level, we comprehensively had to determine athletic performance levels. In this study, the following method was used to determine athletic performance. Based on the previous athletic performance records in the official competitions of the sports, we divided the subjects into 2 athletic performance levels: good athletic performance (GAP) and poor athletic performance (PAP). GAP included subjects who took part in Olympic games and/or international athletic competitions in each sport.
The data are expressed as mean ± SD. The physical characteristics, the CSA values of trunk muscles, and trunk muscle strength in both the collegiate wrestlers and judokas were compared by using the unpaired Student t-test. Likewise, for athletic performance of each sport, we compared those with GAP with those with PAP by using the unpaired Student t-test. All statistical analyses were performed using SPSS (11.0J; SPSS Japan Inc., Tokyo, Japan). The statistical significance level was selected as 5%.
The parameters describing the physical characteristics of the collegiate wrestlers and judokas are listed in Table 1. The sport history of the judokas was significantly longer than that of the wrestlers (P < 0.01). No significant differences were observed between the wrestlers and judokas with regard to the other parameters.
Table 2 shows physical characteristics of the GAP and PAP in wrestling and judo. The height of those with GAP in judokas was significantly greater than that of those with PAP (P < 0.05).
Cross-sectional Areas of Trunk Muscles
The absolute and relative CSAs of the trunk muscles of the collegiate wrestlers and judokas are compared in Table 3 to examine any differences that may be explained by the sport-specific characteristics of the sports practiced. With regard to both the absolute and relative values, a significantly larger RA was observed in the wrestlers (P < 0.05 and P < 0.05, absolute and relative, respectively). The absolute and relative trunk muscle CSAs of the judokas were statistically revealed to be significantly greater than those of the wrestlers (OB [P < 0.05 and P < 0.05, absolute and relative, respectively] and QL [P < 0.01 and P < 0.01, absolute and relative, respectively]).
In Table 4, the relative trunk muscle CSAs on athletic performance levels are presented in each sport. The CSA parameters almost did not show significant differences except for RA in judokas (P < 0.05).
Trunk Muscle Strength
To evaluate trunk muscle strength, isokinetic trunk muscle strength was measured in the collegiate wrestlers and judokas. Table 5 (extensor strength) and 6 (flexor strength) provide the differences between the wrestlers and judokas with regard to the absolute and relative values of the isokinetic trunk muscle strength parameters.
In the absolute trunk extensor strength (Table 5), the values of peak torque at 60°·s−1 (P < 0.01); work at 60°·s−1, 90°·s−1, and 120°·s−1 (P < 0.01, P < 0.01, and P < 0.05, respectively); and average torque at 60°·s−1 and 90°·s−1 (P < 0.05 and P < 0.05, respectively) were significantly higher in the wrestlers than in the judokas. In the relative trunk extensor strength (Table 5), the values of peak torque at 60°·s−1 (P < 0.01); work at 60°·s−1, 90°·s−1, and 120°·s−1 (P < 0.01, P < 0.01, and P < 0.01, respectively); and average torque at 60°·s−1 and 90°·s−1 (P < 0.05 and P < 0.05, respectively) were significantly higher in the wrestlers than in the judokas.
In the absolute and relative trunk flexor strength (Table 6), the values of peak torque at 120°·s−1 (P < 0.01) and work at 90°·s−1 and 120°·s−1 (P < 0.05 and P < 0.01, respectively) were significantly higher in the wrestlers than in the judokas. With regard to the absolute and relative values of the trunk flexor and extensor strength parameters, most of the values were higher in the wrestlers than in the judokas.
The relative trunk extensor (Table 7) and flexor (Table 8) strength parameters on athletic performance levels were compared between GAP and PAP in each sport. However, in this study, none of the parameters were significantly different.
This study examined the sport-specific characteristics of the trunk muscles of collegiate wrestlers and judokas by comparing the CSAs of trunk muscles and trunk muscle strength. We observed obvious differences in the sport-specific characteristics of the CSAs of trunk muscles and trunk muscle strength (wrestlers for flexion and extension motions, judokas for rotation and lateral flexion motions), although both competitive sports are similar with regard to certain aspects. Obviously, these results do not imply that wrestlers are never required to perform trunk rotation and lateral flexion motions; similarly, they do not imply that judokas never perform trunk flexion and extension motions. We believe that this study is the first to describe the sport-specific characteristics of the CSAs of trunk muscles and trunk muscle strength, even if the 2 sports are similar.
Moreover, to confirm relationships between athletic performance and the CSAs and muscular strength of trunk muscles, the CSAs of trunk muscles and trunk muscle strength were compared between athletic performance levels of both the wrestlers and judokas. However, no significant differences have been observed in athletic performance levels in this study, except for the CSA of the RA in judokas.
Significant differences were observed between the 2 sports with regard to the absolute and relative CSA values of the RA, OB, and QL muscles in the athletes. Further, we discuss the correlation between the specific trunk motions involved in these sports and the 3 trunk muscle groups (RA, OB, and QL).
In the present study, the CSA of the RA was significantly higher in the wrestlers than in the judokas. Recently, Kubo et al. (12) also reported that the CSA of the RA in elite wrestlers was significantly larger than that in elite junior wrestlers. Hinson (6) demonstrated that the primary action of the RA is spinal flexion. It is very important for wrestlers to strengthen their trunk flexor and extensor motions because wrestlers are often required to assume the low posture that is unique to wrestling and are also required to lift an opponent during practice and competitions. While assuming this unique low posture and lifting an opponent, wrestlers are required to firmly stabilize their trunk region; because of this, wrestlers should require a larger RA since the RA muscles are antagonistic to the trunk extensor muscles. Therefore, we believe that the RA indicated a significant difference between the wrestlers and judokas by the sport-specific characteristic.
The CSA of the OB was significantly larger in the judokas than in the wrestlers. In this study, the OB muscles were defined as comprising the internal and external obliques as well as the transversus abdominis muscles; these muscles were divided by a poorly defined border. In addition, the actions of both the internal and the external oblique muscles will result in the flexion of the vertebral column. However, one side of the activated muscles rotates and laterally flexes the vertebral column; the internal oblique rotates the vertebral column to the same side, and the external oblique rotates it to the opposite side (6). There is a strong possibility that judokas have greater trunk rotator and lateral flexor strength because it was observed that the OB in these athletes have a greater CSA and less trunk flexor strength. Some studies have revealed that the trunk rotator and lateral flexor strengths are crucial for judokas (5,9,21). Judokas demonstrate the remarkable ability to throw an opponent during practice and competitions. This throwing motion would aid in the development of trunk rotator and lateral flexor muscles and improve their strength, which play an important role in the throwing motion. Further, when an opponent grasps the judo uniform referred to as judogis, particularly its collar and sleeves, the judoka attempts to force the opponent to release the judogis by simultaneously exerting various trunk rotation and lateral flexion motions in order to prevent the opponent from gaining advantage. If the judoka does not permit the opponent to grasp the judogis, it is possible that the judoka will not be thrown. It is therefore expected that the CSA of the OB may be larger in judokas than in wrestlers because judokas use the trunk rotation and lateral flexion motions. In both the standing and groundwork techniques, trunk rotations and lateral flexions are very important for judokas.
Moreover, the judokas were observed to have a QL with a significantly larger CSA. One side of the QL laterally flexes the vertebral column, and both sides stabilize the vertebral column (6). The significantly larger CSA observed in judokas may be attributed to the considerable use of lateral flexion, and the importance of QL is similar to that of OB, as described above. Thus, the specific characteristic of the trunk muscle was observed to differ between the collegiate judokas and wrestlers. The sport-specific trunk motions involved in each sport can induce muscular hypertrophy in the corresponding trunk muscles.
Next, we discuss the fact that our findings indicate the need for wrestlers to strengthen their sagittal movements such as trunk flexion and extension motions, while judokas are required to strengthen their trunk rotation and lateral flexion motions. In the present study, the absolute and relative values of the trunk muscle strength parameters of competitive collegiate wrestlers and judokas were compared. The results indicated that the wrestlers had significantly higher isokinetic trunk extensor and flexor strength than the judokas (in extension, peak torque at 60°·s−1; work at 60°·s−1, 90°·s−1, and 120°·s−1; and average torque at 60°·s−1 and 90°·s−1 and in flexion, peak torque at 120°·s−1 and average torque at 90°·s−1 and 120°·s−1). The other values of trunk flexor and extensor strength were also higher in the wrestlers; however, these values were not significantly different from those of the judokas. We considered that the wrestlers would require greater trunk flexor and extensor strength than the judokas. One reason for this greater requirement is that wrestlers frequently tackle and lift an opponent in competitions and during practice in order to score points, and the aim of the sport is to pin down an opponent. Additionally, wrestlers have nothing to grasp such as judogis as the judokas have. It is necessary for wrestlers to closely grapple with an opponent. Consequently, the approach to the opponent would necessitate greater trunk flexor and extensor strength in wrestlers.
The trunk muscle strength parameters in this study were examined using the isokinetic trunk flexor and extensor strength with peak torque, work, average torque, and average power. This study particularly focused on trunk muscle strength during sagittal movements. However, only a few previous studies have shown that trunk rotator and lateral flexor strengths are some of important elements with regard to trunk muscle strength in judokas and are also key points from the viewpoint of obtaining a flexible body (5,9,21). It appears that the trunk rotator and lateral flexor strengths observed in a competitive judo are relative to those observed in judo performance and play an important role in the standing and groundwork judo techniques. This was confirmed by MRI performed in this study. In a future study, we would like to examine the trunk rotator and lateral flexor strength in athletes, particularly in judokas and wrestlers. The simultaneous testing of trunk rotator and lateral flexor strength and their correlation to low back pain are very interesting.
On athletic performance, the current study individually determined athletic performance level (GAP and PAP) using the previous official competition records of the sports. Since athletic performance is influenced by many factors included physiology, tactics, and psychology (19), it is very complex and difficult to estimate athletic performance. We discuss individual athletic performance below.
The athletic performance of this study was evaluated using the previous athletic performance records of the official competitions of the sports. Although the evaluation may be not a complete method at the present time, it is thought that the evaluation can reflect athletic performance well. It was reported that many researchers evaluated the physical ability of different group levels on various sports (2,4,12,14). It is certainly important for athletes and coaches to examine physical ability at different group levels. However, there are few reports of athletic performance at the individual level. It is possible that our new approach to assessing athletic performance undergoes drastic changes to the future assessment of athletic performance, especially in individual athletic events.
By comparing the 2 sports, we found high values for the RA in wrestlers and the QL in judokas. With regard to athletic performance levels, none of the CSAs in wrestlers were significantly different. Moreover, only RA and GAP in judokas showed significantly lower values than that of PAP. Although the excessive strengthening of RA might not be necessary for judokas, we think that this possibility is low. In addition, analysis of trunk muscle strength showed no significant difference between GAP and PAP. Above all, there was no apparent characteristic in trunk muscles at the point of athletic performance in the current study.
The judokas with PAP showed a significantly greater CSA only of the RA. None of significant differences were observed in the other CSAs, including trunk extensor and flexor strength parameters. As stated above, the RA in the judokas might prove that a relationship of CSA with athletic performance was relatively low in judo particularly. This is consistent with the result in this study showing that the QL is a characteristic trunk muscle in judokas.
Also, long sport history and high performance level may be the reasons that we cannot observe the difference in trunk muscles between the GAP and PAP in the 2 sports. When athletes practice their sports a lot, they gradually become skillful. Improving their skill may not depend on their muscle strength. In that sense, to develop muscular strength is essential but not enough for performance improvement. We consider this one of the reasons that trunk muscles did not show a significant difference depending on athletic performance levels. In other words, our findings might suggest that larger trunk CSAs are essential but minimal requirements for elite athletes.
In conclusion, the present study examined the sport-specific characteristics of the trunk muscles of collegiate wrestlers and judokas. The CSAs of the trunk muscles in the wrestlers and judokas were significantly different. The results of this study thus indicate that even in these 2 similar competitive sports-wrestling and judo-different sport-specific characteristics are exhibited by the trunk muscles of the athletes. We cannot confirm that the CSAs of trunk muscles and trunk muscle strength significantly differ on athletic performance levels in this study.
The findings of the present study are that obvious differences are present in the sport-specific characteristics of the CSAs of trunk muscles and trunk muscle strength between the collegiate athletes of the 2 similar sports, namely, wrestling and judo. On the sports scene, it is very important to adopt sport-specific training programs and to develop the trunk muscles of athletes involved in these 2 sports. Wrestlers should concentrate heavily on trunk flexion and extension exercises and judokas should concentrate on trunk rotation and lateral flexion exercises. It will be especially necessary for beginners and inexperienced athletes to perform their respective training programs. From the obtained results, we believe that larger trunk CSAs are minimal requirements for elite athletes.
1. Andersson, E, Sward, L, and Thorstensson, A. Trunk muscle strength in athletes. Med Sci Sports Exerc
20: 587-593, 1988.
2. Baker, D. Differences in strength and power among junior-high, senior-high, college-aged, and elite professional rugby league players. J Strength Cond Res
16: 581-585, 2002.
3. Basmajian, JV and De Luca, CJ. Muscles alive: their functions revealed by electromyography. Basmajian, JV and DeLuca, CJ, eds. In: Back
. Baltimore: Williams & Wilkins, 1985. pp. 354-366.
4. Cometti, G, Maffiuletti, NA, Pousson, M, Chatard, JC, and Maffulli, N. Isokinetic strength and anaerobic power of elite, subelite and amateur French soccer players. Int J Sports Med
22: 45-51, 2001.
5. Hattori, M, Imaizumi, T, and Suzuki, N. [Morphological characteristics of trunk muscle groups in Japanese female elite judo athletes] [in Japanese]. Jpn J Phys Fitness Sports Med
42: 485-494, 1993.
6. Hinson, MM. Kinesiology. In: The Spine and Pelvic Girdle
. Dubuque, IA: William C, Brown Company Publisher, 1981. pp. 132-165.
7. Horswill, CA. Applied physiology of amateur wrestling. Sports Med
14: 114-143, 1992.
8. Housh, TJ, Johnson, GO, Hughes, RA, Housh, DJ, Hughes, RJ, Fry, AS, Kenney, KB, and Cisar, CJ. Isokinetic strength and body composition of high school wrestlers across age. Med Sci Sports Exerc
21: 105-109, 1989.
9. Imamura, R and Johnson, B. A kinematic analysis of a judo leg sweep: major outer leg leap-osoto-gari. Sports Biomech
2: 191-201, 2003.
10. Iwai, K, Nakazato, K, Irie, K, Fujimoto, H, and Nakajima, H. [Physical characteristics of university wrestlers with low back pain] [in Japanese]. Jpn J Phys Fitness Sports Med
51: 423-436, 2002.
11. Iwai, K, Nakazato, K, Irie, K, Fujimoto, H, and Nakajima, H. Trunk muscle strength and disability level of low back pain in collegiate wrestlers. Med Sci Sports Exerc
36: 1296-1300, 2004.
12. Kubo, J, Ohta, A, Takahashi, H, Kukidome, T, and Funato, K. The development of trunk muscles in male wrestlers assessed by magnetic resonance imaging. J Strength Cond Res
21: 1251-1254, 2007.
13. Kuroki, T, Tajima, N, and Higuchi, J. Muscle strength characteristics of knee and trunk in judokas. Jpn J Orthop Sports Med
14: 418-422, 1994.
14. Mayhew, JL, Prinster, JL, Ware, JS, Zimmer, DL, Arabas, JR, and Bemben, MG. Muscular endurance repetitions to predict bench press strength in men of different training levels. J Sports Med Phys Fitness
35: 108-113, 1995.
15. Nadler, SF, Wu, KD, Galski, T, and Feinberg, JH. Low back pain in college athletes. A prospective study correlating lower extremity overuse or acquired ligamentous laxity with low back pain. Spine
23: 828-833, 1998.
16. Parkin, S, Nowicky, AV, Rutherford, OM, and McGregor, AH. Do oarsmen have asymmetries in the strength of their back and leg muscles? J Sports Sci
19: 521-526, 2001.
17. Reid, JG, Livingston, LA, and Pearsall, DJ. The geometry of the psoas muscle as determined by magnetic resonance imaging. Arch Phys Med Rehabil
75: 703-708, 1994.
18. Rezasoltani, A, Ahmadi, A, Nehzate-Khoshroh, M, Forohideh, F, and Ylinen, J. Cervical muscle strength measurement in two groups of elite Greco-Roman and free style wrestlers and a group of non-athletic subjects. Br J Sports Med
39: 440-443, 2005.
19. Smith, DJ. A framework for understanding the training process leading to elite performance. Sports Med
20. Sward, L, Svensson, M, and Zetterberg, C. Isometric muscle strength and quantitative electromyography of back muscles in wrestlers and tennis players. Am J Sports Med
18: 382-386, 1990.
21. Takahashi, R. Power training for judo: plyometric training with medicine balls. Natl Strength Cond Assoc J
14: 66-71, 1992.
22. Tracy, MF, Gibson, MJ, Szypryt, EP, Rutherford, A, and Corlett, EN. The geometry of the muscles of the lumbar spine determined by magnetic resonance imaging. Spine
14: 186-193, 1989.
23. Weir, JP, Housh, TJ, Johnson, GO, Housh, DJ, and Ebersole, KT. Allometric scaling of isokinetic peak torque: the Nebraska Wrestling Study. Eur J Appl Physiol Occup Physiol
80: 240-248, 1999.
24. Wood, S, Pearsall, DJ, Ross, R, and Reid, JG. Trunk muscle parameters determined from MRI for lean to obese males. Clin Biomech (Bristol, Avon)
11: 139-144, 1996.
25. Wroble, RR and Albright, JP. Neck and low back injuries in wrestling. Clin Sports Med
5: 295-325, 1986.
26. Yamamura, C, Zushi, S, Takata, K, Ishiko, T, Matsui, N, and Kitagawa, K. Physiological characteristics of well-trained synchronized swimmers in relation to performance scores. Int J Sports Med
20: 246-251, 1999.