Secondary Logo

Journal Logo

Original Research

Bilateral Asymmetry in Joint Torque During Squat Exercise Performed by Long Jumpers

Kobayashi, Yuji; Kubo, Junjiro; Matsuo, Akifumi; Matsubayashi, Takeo; Kobayashi, Kando; Ishii, Naokata

Author Information
Journal of Strength and Conditioning Research: October 2010 - Volume 24 - Issue 10 - p 2826-2830
doi: 10.1519/JSC.0b013e3181c64387
  • Free



Squat exercises are often regarded as an almost symmetric task with little difference in the movement between the 2 lower limbs (5,6). They are frequently performed to symmetrically strengthen both lower limbs. Several studies have been performed on the bilateral asymmetry of the squat exercise (7,12,14) and other tasks involving putative symmetric movement, including movement from the seated to the standing position (9), landing (16), lifting a crate (10), external impact loading (3), and vertical jumping (1,12). Results of previous studies indicated that bilateral asymmetry negatively affected performance in middle-distance running (11) and was a factor associated with injury (4,8,13,18); therefore, it is better to avoid bilateral asymmetry as much as possible.

It is important to vary the load during resistance training. For example, to acquire hypertrophy as a training adaptation, it is necessary to increase the load to at least 67% of 1 repetition maximum (RM) or more (2). However, very few studies have investigated the relationship between load and bilateral asymmetry in the squat exercise. This relationship has been studied only in recreationally trained men and women (7), and even under light loads, bilateral differences occur among several kinematic and kinetic parameters. In addition, increasing the external load did not necessarily lead to increases in the bilateral differences in any of the variables examined in the previous study (7). However, only one study on female softball players (12) has been conducted in athletes under a heavy loading condition (80% of 1RM); therefore, it is unclear whether bilateral differences are likely to occur when athletes squat under varying loading conditions.

Many sports involve asymmetric movements. The long jump is a typical example of such a sport because it requires a large ground reaction force (GRF) and involves large joint torques on the takeoff leg (TL) at the moment of jumping (17). Therefore, during squats, long jumpers might lean on the TL that is accustomed to generating stronger torques, which might lead to bilateral asymmetry. Moreover, this bilateral asymmetry might increase under heavy loading condition because additional loading requires greater torque generation.

Thus, in this study, we aimed to investigate the kinematics and kinetics of the bilateral lower limbs in athletes when they performed squats with different loads. Bilateral GRFs, joint angles, and torques were used as parameters of comparison between the limbs. We hypothesized that sports-specific training involving asymmetric movement would cause alterations in motor control while performing the squat exercise. In particular, long jumpers are expected to have larger joint torques in TL than in the non-takeoff leg (NTL) during squats. We also hypothesized that additional loading increases bilateral asymmetry.


Experimental Approach to the Problem

Long jumpers performed 3 repetitions of the squat exercise with 3 different loads. During the exercise, their body positions and GRFs were recorded using a motion capture system and 2 force platforms. The data obtained were used to calculate joint kinematics and kinetics of TL and NTL during squats using inverse dynamics. Each parameter was compared between TL and NTL.


Eighteen male university students (height: 174.6 ± 6.5 cm [mean ± SD]; weight: 67.3 ± 4.7 kg; and age: 21.6 ± 1.7 years) participated in this study. They were regional level athletes trained for long jump and had participated in local intercollegiate games in Japan. All the subjects had performed systematic resistance training based on the traditional periodization model (19) at least twice per week for at least 3 years, and their daily training programs included the barbell squat exercise. During training, the subjects assessed their squat movements using a mirror. None of the subjects reported a history of severe musculoskeletal disorder or injuries (such as bone fracture, ligament damage, and severe sprain) in the lower extremities. According to the guidelines of the Ethics Committee for Human Experiments, Japan Institute of Sports Sciences, the subjects were informed of the methods, purposes, and potential risks of the study, and their written informed consents for participation in the study were obtained.


Three days before the experiment, the value of 3RM (114 ± 27 kg) was determined for each subject. On the day of the experiment, the subjects performed warm-up exercises, and the depth (lowermost position) of the squat and the foot position assumed during the squat were determined for each subject. The squat depth was determined as the position where both thighs were parallel to the floor when the subjects performed the squat with a bar weighing 20.0 kg. Then, the position of the bar was recorded on poles positioned near the subjects. Their most familiar stance width and foot abduction angle was used as the foot position. The foot abduction angle was defined as the angle between the line joining the head of the second metatarsal to the heel of each foot and the axis perpendicular to the line joining both heels. The mean stance width was 112 ± 17% of the shoulder width, and the mean foot abduction angles in TL and NTL were 15 ± 4° and 15 ± 5°, respectively. The subjects performed 3 repetitions of the squat exercise at 50% of 3 RM at the same speed of movement with which they normally performed their training. Thereafter, the load was increased to 70% and then to 90% of 3RM. The order of loading was not randomized to avoid the risk for injury associated with immediate loading of a heavy weight.

Kinematic and Kinetic Data Collection

Ground Reaction Force Measurement

Two force platforms (Kistler, 9287B, Winterthur, Switzerland) were independently used to measure the GRFs under each foot during the squat exercises. The GRFs were recorded at 600 Hz and divided into their vertical and horizontal components (vGRF and hGRF).

Motion Capture System

Reflective markers were placed on anatomical landmarks (second metatarsal head; lateral malleolus; heel; and the lateral aspects of the knee joint, greater trochanter, anterior superior iliac spine, and posterior superior iliac spine) of the body. This enabled the body to be modeled as a linked chain from the pelvis, thigh, shank, and foot. The movements of the subjects during the squat exercise were recorded using a 12-camera motion analysis system (Oxford Metrics, Vicon, United Kingdom) at 60 Hz and synchronized with the force platform data.

Data Calculation

Sagittal plane kinematics and kinetics of the ankle, knee, and hip during the squat exercise were calculated for each subject for each trial using the inverse dynamics approach applied in Vicon's Plug-In-Gait software (Oxford Metrics, Vicon, United Kingdom). All joint angles were termed the flexion angle. Joint torques were normalized to the body mass of each subject and expressed as extensor torque.

Statistical Analyses

The values obtained for peak vGRF and hGRF, maximal flexion angle, and peak extensor torque at the hip, knee, and ankle were averaged among 3 repetitions of the squat exercise with each load. These variables were expressed in terms of mean and SD. The reliability of the variables used in this experiment was assessed using the intraclass correlation coefficient (ICC) among 3 repetitions with each load. A 2-way analysis of variance (ANOVA) with repeated measures was used to examine the effect of loads and limbs on these variables. Post hoc analyses with paired t-tests were used to compare the differences between TL and NTL. All statistical analyses were performed using a statistical software package (SPSS for Windows, version 16.0; SPSS Inc, Chicago, IL, USA).


Intraclass Correlation Coefficient

The ICCs of the parameters in this experiment were within the range of 0.728-0.997, indicating that the parameters were sufficiently reliable.

Ground Reaction Force

A 2-way ANOVA with repeated measures for peak vGRF and hGRF revealed significant effects of the loads (p < 0.001). However, no other significant effects were detected (limb, p = 0.515 and 0.438, respectively; interaction, p = 0.771 and 0.645, respectively). Table 1 shows the peak vGRF and hGRF produced during squat at 50, 70, and 90% of 3RM.

Table 1
Table 1:
Comparison of the peak vGRF and hGRF (N, mean ± SD) between the TL and NTL when performing squats at 50, 70, and 90% of 3 RM.*

Joint Angle

A 2-way ANOVA with repeated measures for the maximal flexion angle at the hip revealed significant effects of loads and limbs (p = 0.049 and 0.031, respectively), but interaction effects were not significant (p = 0.445). A 2-way ANOVA with repeated measures for the maximum flexion angle at the knee revealed significant effects of loads (p = 0.027); however, no other significant effects were observed (limb, p = 0.231; interaction, p = 0.846). A 2-way ANOVA with repeated measures for the maximal flexion angle at the ankle did not reveal any significant effects (load, p = 0.508; limb, p = 0.768; interaction, p = 0.847). Table 2 shows the maximum flexion angle at the hip, knee, and ankle joints during the squat exercises performed with loads of 50, 70, and 90 of 3 RM.

Table 2
Table 2:
Comparison of the maximal flexion angle (degrees, mean ± SD) between the TL and NTL around the hip, knee, and ankle when performing squats at 50, 70, and 90% of 3RM.*

Joint Torque

A 2-way ANOVA with repeated measures for peak hip joint torque revealed significant main effects (load, p < 0.001; limb, p = 0.022), although the interaction effects were not significant (p = 0.289). A 2-way ANOVA with repeated measures for peak knee joint torque revealed significant effects of loads (p < 0.001), but no other significant effects were present (limb, p = 0.165; interaction, p = 0.279). A 2-way ANOVA with repeated measures for peak ankle joint torque revealed significant main and interaction effects (load, p < 0.001; limb, p = 0.045; and interaction, p = 0.013). Post hoc analyses for the ankle joints revealed that the peak extensor torque generated during the squat performed with 90% of 3RM was greater in TL than in NTL (p = 0.011). However, no significant differences were present in the peak joint torque produced at the ankle between TL and NTL at 50% and 70% of 3 RM (p = 0.328 and 0.122, respectively). Table 3 shows peak joint torques at the hip, knee, and ankle produced during the squat exercise performed with 50, 70, and 90% of 3 RM.

Table 3
Table 3:
Comparison between peak extensor torques (N·m·kg−1, mean ± SD) in the TL and NTL around the hip, knee, and ankle during squats performed at 50, 70, and 90% of 3 RM.*


The present results demonstrated that there were significant differences in the maximal joint angle and peak joint torques at the hip between TL and NTL under all loading conditions. In addition, ankle-joint torques produced at 90% of 3RM were significantly different between TL and NTL. However, despite the differences in joint torque, there were no differences in GRF during any of the loading conditions.

We hypothesized that sports-specific training involving asymmetric movement would cause alterations in the contributions of TL and NTL while performing squats. Our results for hip joint torque, which revealed differences between TL and NTL under all loading conditions, appear to support our initial hypothesis though no interaction between the load and the limb disagreed with our second hypothesis that additional loading increases bilateral differences. Previous studies on movement from the seated to the standing position (9) and squat (7), which examined lower limb joint torques using inverse dynamics, agree with the present study in that there was a bilateral difference in hip joint torque. However, in the present study, a significant difference was not observed in the torque produced at the knee between TL and NTL under any loading condition, and thus, the hypothesis was not supported. In a previous study on the squat exercise (7), bilateral differences were observed in the knee. On the contrary, there was no bilateral difference in the knee in a previous study on movement from the seated to the standing position (9) and the present study. Bilateral differences were observed at the ankle joint under 90% of 3RM in the present study. There was also an interaction between the load and the side in a previous study performed under 4 loading conditions (7); however, no bilateral differences were observed for the heaviest load (100% of 3RM), though bilateral differences occurred under the other 3 loading conditions (25, 50, and 75% of 3RM) (7).

The generalized motor program hypothesis (15) referred to in a previous study (7) suggests that bilateral symmetry remains unaffected by loading conditions. Because bilateral symmetry in the ankle torque was changed depending on the load, the present study was not in agreement with this hypothesis, whereas the hip and knee results of the present study, and that of the previous study (7), supported this hypothesis.

The results of the present study were inconsistent with 2 intuitive assumptions. The first assumption is that the bilateral difference in GRF can influence the bilateral difference in torque, and that larger torque is produced at the side with a larger GRF. The second assumption is that a large flexion angle may indicate a larger torque. However, the present study showed bilateral differences in hip and ankle joint torques between TL and NTL despite the fact that the GRF values between TL and NTL did not differ significantly. Moreover, the peak torque at the hip in TL was larger than that in NTL, whereas the maximal flexion angle at the hip in TL was smaller than that in NTL. The multiple-joint characteristics of movements such as squat exercises may allow trainees to use substitution patterns that shift the effort from the targeted muscle group to another muscle group (14). In the previous study, when GRF of one limb is the same as that of the contralateral limb, joint torques in lower limbs are not necessarily the same on both sides (14) and as in the present study. Results in the previous study indicated that there was substitution between both limbs (14); however, this could be insufficient to explain the present results. Not only lower limbs but also the body trunk should be taken into consideration. Bilateral asymmetry in hip-joint angle without that in knee and ankle joint angle suggest trunk sway in the frontal direction and left-right asymmetry in trunk kinetics. According to this suggestion, we assumed that the torque produced at lumbar or thoracic spine or both could compensate for an imbalance in the bilateral lower limb kinetics. If this assumption is true, attention should be paid to the imbalance of the kinetics of the legs because an imbalance in trunk kinetics may cause low-back injury. However, this hypothesis would require further investigation.

At the ankle joint, bilateral difference in the torque was observed when the subjects performed squats only under a load of 90% of 3RM, suggesting that additional loading likely influenced the bilateral symmetry of the ankle joint torque. Care should be taken to maintain the bilateral symmetry of the ankle joint torque to prevent injury when a load added. However, because the change was not shown in the kinematics, the clue to correct this asymmetry is not found from the results of the present study.

A limitation of this study was the difficulty of precisely controlling the speed of the squat movement, and differences in the speed of movement among subjects could have influenced the results. Moreover, we did not measure bilateral differences in the muscle strength, tightness, and bone alignment in this study; therefore, the influences of these factors remain unclear. In addition, further research regarding other sports is needed to determine whether our results can be applied to athletes performing other sports.

In conclusion, bilateral asymmetry in the hip-joint angle and torque occurred when squats were performed under all loading conditions. There was a bilateral difference in ankle joint torque when loads became heavy. These results should be considered when attempting to decrease the risk of injury.

Practical Applications

To maintain joint torques in both legs equally during the squat, placing the weight equally on both feet might not be necessarily sufficient because a bilateral difference was observed in joint torque despite no bilateral difference in GRF between the sides in the present study. Moreover, bilateral difference of joint torque in both limbs might be compensated for by trunk kinetics (imbalance of trunk kinetics), and this may cause low back injury. These should be considered when attempting to decrease the risk of injury.


The present study was conducted as a part of research projects of the Japan Institute of Sports Sciences. We wish to express our thanks to everyone who supported us during the experiments and discussed about the results.


1. Bobbert, MF, de Graaf, WW, Jonk, JN, and Casius, LJ. Explanation of the bilateral deficit in human vertical squat jumping. J Appl Physiol 100: 493-499, 2006.
2. Campos, GE, Luecke, TJ, Wendeln, HK, Toma, K, Hagerman, FC, Murray, TF, Ragg, KE, Ratamess, NA, Kraemer, WJ, and Staron, RS. Muscular adaptations in response to three different resistance-training regimens: Specificity of repetition maximum training zones. Eur J Appl Physiol 88: 50-60, 2002.
3. Chavet, P, Lafortune, MA, and Gray, JR. Asymmetry of lower extremity responses to external impact loading. Hum Mov Sci 16: 391-406, 1997.
4. Croisier, JL, Forthomme, B, Namurois, MH, Vanderthommen, M, and Crielaard, JM. Hamstring muscle strain recurrence and strength performance disorders. Am J Sport Med 30: 199-203, 2002.
5. Escamilla, RF. Knee biomechanics of the dynamic squat exercise. Med Sci Sports Exerc 33: 127-141, 2001.
6. Flanagan, S, Salem, GJ, Wang, MY, Sanker, SE, and Greendale, GA. Squatting exercises in older adults: Kinematic and kinetic comparisons. Med Sci Sports Exerc 35: 635-643, 2003.
7. Flanagan, SP and Salem, GJ. Bilateral differences in the net joint torques during the squat exercise. J Strength Cond Res 21: 1220-1226, 2007.
8. Knapik, JJ, Bauman, CL, Jones, BH, Harris, JM, and Vaughan, L. Preseason strength and flexibility imbalances associated with athletic injuries in female collegiate athletes. Am J Sports Med 19: 76-81, 1991.
9. Lundin, TM, Grabiner, MD, and Jahnigen, DW. On the assumption of bilateral lower-extremity joint moment symmetry during the sit-to-stand task. J Biomech 28: 109-112, 1995.
10. Maines, JM and Reiser, RF. Ground reaction force bilateral asymmetries during submaximal sagittal plane lifting from the floor. Int J Ind Ergonom 36: 109-117, 2006.
11. Manning, JT and Pickup, LJ. Symmetry and performance in middle distance runners. Int J Sports Med 19: 205-209, 1998.
12. Newton, RU, Gerber, A, Nimphius, S, Shim, JK, Doan, BK, Robertson, M, Pearson, DR, Craig, BW, Häkkinen, K, and Kraemer, WJ. Determination of functional strength imbalance of the lower extremities. J Strength Cond Res 20: 971-977, 2006.
13. Orchard, J, Marsden, J, Lord, S, and Garlick, D. Preseason hamstring muscle weakness associated with hamstring muscle injury in Australian footballers. Am J Sports Med 25: 81-85, 1997.
14. Salem, GJ, Salinas, R, and Harding, FV. Bilateral kinematic and kinetic analysis of the squat exercise after anterior cruciate ligament reconstruction. Arch Phys Med Rehabil 84: 1211-1216, 2003.
15. Schmidt, RA. Schema theory of discrete motor skill learning. Psychol Rev 82: 225-260, 1975.
16. Schot, PK, Bates, BT, and Dufek, JS. Bilateral performance symmetry during drop landing - a kinetic-analysis. Med Sci Sport Exerc 26: 1153-1159, 1994.
17. Stefanyshyn, DJ and Nigg, BM. Contribution of the lower extremity joints to mechanical energy in running vertical jumps and running long jumps. J Sport Sci 16: 177-186, 1998.
18. Tyler, TF, Nicholas, SJ, Campbell, RJ, and McHugh, MP. The association of hip strength and flexibility with the incidence of adductor muscle strains in professional ice hockey players. Am J Sports Med 29: 124-128, 2001.
19. Wathen, D, Baechle, TR, and Earle, RW. Periodization. In: Essentials of Strength Training and Conditioning. Baechle, TR and Earle, RW, eds. Champaign, IL: Human Kinetics, 2008. pp. 507-522.

takeoff leg; kinematics; kinetics; sports specificity; load

Copyright © 2010 by the National Strength & Conditioning Association.