Secondary Logo

Journal Logo

Original Research

A Biomechanical Comparison of the Traditional Squat, Powerlifting Squat, and Box Squat

Swinton, Paul A.1; Lloyd, Ray2; Keogh, Justin W. L.3,4; Agouris, Ioannis1; Stewart, Arthur D.5

Author Information
Journal of Strength and Conditioning Research: July 2012 - Volume 26 - Issue 7 - p 1805-1816
doi: 10.1519/JSC.0b013e3182577067
  • Free



The squat is one of the most frequently used resistance exercises for strength development in both athletic and rehabilitation settings. As a result of its widespread use, the exercise has been the focus of a large number of biomechanical studies (10–12,19,21,23,24,28). The results present the squat as a complex movement that requires coordinated actions of the torso and all major joints of the lower extremities (10,20). Furthermore, this complexity enables individuals to select different movement strategies to perform the exercise. From a performance enhancement and injury risk perspective, it is commonly recommended that movement strategies used to perform the squat should minimize anterior displacement of the knee (5). This recommendation is based on the reasoning that maintenance of a near vertical shin position during the squat reduces internal forces at the knee and emphasizes recruitment of the hip extensor muscles (5,6). The first study to investigate the effects of controlling anterior knee displacement during the squat was conducted by Fry et al. (12). The investigators measured joint torques produced at the hip and knee when squats were performed under 2 conditions with differing amounts of anterior knee displacement. During the first condition, the subjects were permitted to displace the knee beyond the toes, whereas during the second condition displacement was restricted by placing a vertical board at the subjects' feet. The results showed that creating a more vertical shin position by restricting anterior displacement decreased torque at the knee while concomitantly increasing torque at the hip. Fry et al. (12) also reported that restricting anterior displacement of the knee created a more horizontal torso position, which suggested that greater shear forces were developed at the lumbar spine. The authors proposed that a more horizontal posture was adopted to compensate for changes in positioning of the lower leg and maintain the system center of mass (COM) over the base of support (12). The results obtained by Fry et al. (12) have caused some to propose that restricting anterior displacement of the knee during squatting may create potentially injurious forces at the lower back (6,12).

The intention to restrict anterior displacement of the knee and maintain a near vertical shin position is a key feature of the movement strategies used by powerlifters to perform the squat. To achieve this posture, many powerlifters adopt a wide stance and focus on moving the hips posteriorly during the descent phase of the movement. In practical settings, this movement strategy is often referred to as “sitting back” and is the characterization of what is considered to represent the powerlifting squat (6,14) (Figure 1, middle). In contrast to the result presented by Fry et al. (12), observation of skilled powerlifters suggests that some individuals can squat with relatively upright torso positions while restricting anterior displacement of the knee. At present, it is not fully understood how these individuals successfully perform this task. However, to develop proficiency in the movement, many powerlifters perform the squat onto a box placed behind the lower leg (26) (Figure 1, bottom). The box enables the performer to maximize posterior displacement of the hip and maintain a vertical shin position by acting as a safety device to catch the individual if the COM is moved beyond the base of support. Both the powerlifting squat and the box squat as is commonly known are now popular exercises used by athletes other than powerlifters to develop strength and power (6,19). However, some researchers and practitioners have questioned the safety and effectiveness of both exercises (4,6). To date, only a limited number of studies have quantified biomechanical variables during the powerlifting squat or box squat. Multiple investigators have collected data from squats performed during powerlifting competitions (10,14,21). However, research studies have established that techniques used by individual powerlifters are varied and that some choose not to restrict anterior displacement of the knee (14). Much less information is available regarding the biomechanics of the box squat. McBride et al. (19) compared kinetic and electromyographic data of powerlifters performing the box squat and what was described as a standard squatting movement. The authors reported only minimal differences in peak force and muscle activity measured at the thigh. The experimental protocol used by McBride et al. (19) did not calculate joint specific data or provide kinematic information regarding the movement strategies used by the powerlifters to perform each exercise. Because of the limited information available at present, coaches and athletes are unable to make informed judgments regarding the appropriateness of the powerlifting squat or box squat. Therefore, it was the principal aim of this study to provide a detailed kinematic and kinetic comparison of each exercise with additional analysis of the traditional squat to provide a reference. In fulfilling this aim, the study objectives included data collection for each exercise over a range of loads performed with the intent to overcome the resistance as fast as possible to simulate the training protocols used frequently to develop muscular strength and power.

Figure 1
Figure 1:
Traditional squat (top), powerlifting squat (middle), and box squat (bottom).


Experimental Approach to the Problem

A cross-sectional, repeated measures design was used to quantify and compare kinematics and kinetics of the traditional squat, powerlifting squat, and box squat. The experimental approach provided original information regarding movement strategies used to perform each exercise and comparative data to assist practitioners in exercise instruction and training prescription. The subjects comprised well-trained powerlifters with extensive experience in performing each exercise. Data were collected for each subject over 2 sessions separated by 1 week. Session 1 was performed in the gymnasium and involved 1-repetition maximum (1RM) testing in the squat. Session 2 was performed in the laboratory where subjects performed maximum speed repetitions for each exercise using loads of 30, 50, and 70% of their recorded 1RM. Kinematics and kinetics were analyzed during session 2 only.


Twelve male powerlifters participated in the study (age: 27.2 ± 4.2 years; stature: 180.3 ± 4.8 cm; mass: 100.2 ± 13.1 kg; squat 1RM: 220.2 ± 36.2 kg; resistance training experience: 9.2 ± 3.1 years). All the subjects had a minimum of 3 years of experience performing each exercise. The study was conducted 3 months after a regional competition where the majority of the subjects were nearing the end of a training cycle aimed at matching or exceeding their previous competition performance. The subjects were notified about the potential risks involved and gave their written informed consent to be included. Prior approval was given by the ethical review panel at Robert Gordon University, Aberdeen, United Kingdom.

One-Repetition Maximum Testing

All the subjects chose to perform the squat 1RM test using the powerlifting technique they used in competition. No supportive aids beyond the use of a weightlifting belt were permitted during the test. Based on a 1RM load predicted from performance in recent training sessions, the subjects performed a series of warm-up sets and up to 5 maximum attempts. A minimum of 2 minutes and a maximum of 4 minutes of recovery time were allocated between attempts (2). Within this time frame, the subjects chose to perform the lifts based on their own perception of when they had recovered. All repetitions were performed to a depth where the thighs became parallel with the floor (2). Each attempt was deemed successful if the appropriate depth was reached and the barbell was not lowered at any point during the ascent phase.

Squat Variation Testing

Before performing maximum speed repetitions, the subjects engaged in their own specific warm-up. Generally, this began with 3–5 sets of light squats (e.g., <40% 1RM) for 6–10 repetitions. All the subjects then performed a series of maximum speed repetitions before any data collection. Once suitably prepared, the subjects performed all 3 exercises with loads of 30, 50, and 70% of their predetermined 1RM. One trial comprising 2 repetitions was performed for each load and condition to assess intratrial reliability. The 9 trials were performed in a randomized order with a minimum 2-minute rest period allocated. A longer rest period of up to 4 minutes was made available if the subject felt it necessary to produce a maximum performance. For the traditional squat, the subjects were instructed to allow the knee to travel past the toes during the descent phase. For the powerlifting squat and box squat, the subjects were instructed to move the hip posteriorly and try to maintain as vertical a shin position as possible. During the box squat, the subjects were permitted to displace the COM behind the base of support during the final portion of the descent and were instructed to pause for a minimum of 1 second on the box. Instructions were given to perform the concentric portion of each repetition with maximum effort attempting to lift the load as fast as possible whilst maintaining contact with the ground throughout the movement. For each trial, the repetition that produced the greatest peak barbell velocity was selected for further analysis.

All testing was completed between 17:00 and 20:00 hours to correspond with the powerlifters' regular training times. The subjects followed their individual nutritional practices used before training sessions. Consumption of water (500 ml) was permitted during tests, and the room temperature was maintained between 22 and 25° C. Consistent verbal encouragement was provided during both testing sessions with the subjects frequently reminded to lift each load as fast as possible.

Biomechanical Instrumentation

A marker was placed on each of the following bony landmarks: spinous process of the 7th cervical vertebra, spinous process of the 10th cervical vertebra, suprasternal notch, inferior tip of the xiphoid process, left and right anterior superior iliac spine, left and right lateral femoral epicondycle, left and right lateral malleolus, and left and right heads of the second metatarsal. Additionally, markers were placed on the sacrum midway between the posterior superior iliac spines and bilaterally at midtibia, midfemur, and the calcaneous. The geometric center of the external load was tracked in 3-dimensional space by placing markers at the ends of the barbell and calculating the position of the midpoint. Trials were performed with a separate piezoelectric force platform (Kistler, Type 9281B Kistler Instruments, Winterthur, Switzerland) under each foot, in a capture area defined by a 9-camera motion analysis system (Vicon MX, Vicon Motion Systems, Oxford Metrics, United Kingdom). Marker position and ground reaction force (GRF) data were captured at 200 and 1,200 Hz, respectively.

Data Processing and Reduction

Based on a frequency content analysis of the 3-dimensional coordinate data, marker trajectories were filtered using a digital fourth-order low-pass Butterworth filter with a cut-off frequency of 6 Hz. A 3-dimensional lower-body model (16) and upper body model (13) were used to calculate joint positions and angles of the torso, hip, knee, and ankle and the position of the fifth lumbar vertebra. Linear and angular velocities were calculated by differentiating position data with a Lagrangian 5-point differentiation scheme. Joint moments were calculated using inverse dynamics and anthropometric data with Vicon Nexus 1.7 processing software (Oxford Metrics, Oxford, United Kingdom). The moment arm created by the external resistance was also calculated for each joint. This was computed by measuring the horizontal distance from the geometric center of the barbell to the respective joint centers. Kinematics and kinetics for the hip, knee, and ankle were calculated for both left and right sides and then averaged to obtain single values. Squatting technique was assessed by using quantitative and qualitative means. Quantitatively, the technique was assessed by measuring joint angles during the first frame of the concentric movement. For qualitative analyses, representative joint angle-time curves were selected and compared across techniques. Similar quantitative and qualitative analyses have been used previously to describe techniques used to perform the squat (10,12). Peak power and rate of force development (RFD) were also measured to assess the external performance of each squat. Instantaneous power values were calculated as the product of the vertical GRF and corresponding vertical barbell velocity. The RFD was calculated from the slope of the vertical GRF-time curve extending from the transition between eccentric and concentric phases to the maximum value of the first peak.

Statistical Analyses

Intratrial reliability for each variable analyzed was assessed by intraclass correlation coefficient (ICC). As recommended by Baumgartner (3), ICCs were calculated with a correction factor for number of repetitions performed per trial (n = 2) and number of repetitions used in the criterion score (n = 1). Intratrial reliability for all variables reported was >0.88. Potential differences in kinematic and kinetic variables measured during the squats were analyzed using a 3 × 3 (squat type × load) repeated measures analysis of variance. Significant main effects were further analyzed with Bonferroni adjusted pairwise comparisons. Statistical significance was accepted at p ≤ 0.05. All statistical procedures were performed using the SPSS software package (SPSS, Version 17.0, SPSS Inc., Chicago, IL, USA).


Linear Kinematics

The powerlifting squat and box squat were performed with a significantly wider stance than the traditional squat (89.6 ± 4.9, 92.1 ± 5.1, 48.3 ± 3.8 cm, respectively). Linear displacements of the barbell and joint centers in the anterior-posterior direction revealed differences across techniques (Table 1). The largest effects were noted during the eccentric phase where greater posterior hip displacements and reduced anterior knee displacements occurred during the powerlifting squat and box squat compared with the traditional squat. These differences were reflected in the overall displacement of the system COM. During the eccentric phase, the system COM was displaced anteriorly during the traditional squat and posteriorly during the powerlifting squat and box squat.

Table 1
Table 1:
Anterior-posterior displacements calculated across the eccentric and concentric phases (mean ± SD).

Angular Kinematics

Potential differences in squatting posture were primarily assessed by recording segmental angles during the first frame of the concentric phase. The values were averaged across loads because the external resistance was found to have minimal effect (Table 2). Similar torso angles were obtained for the traditional squat and powerlifting squat. However, at the start of the concentric phase, a significantly more upright torso was recorded for the box squat. Angular differences across the exercises were observed at all 3 joint axes of the hip. The wide stance squats (powerlifting and box) displayed significantly greater abduction angles than the traditional squat. In addition, significantly greater hip flexion and internal rotation was recorded during the powerlifting squat compared with the other exercises. Significant differences were also obtained for the knee and ankle, with greater flexion angles obtained at both joints during the traditional squat.

Table 2
Table 2:
Joint angles at the start of the concentric phase (mean ± SD).

A qualitative assessment of the lifting technique adopted for each exercise was obtained by selecting representative joint angle-time curves. Comparatively homogeneous traces were obtained for the traditional squat (Figure 2). The results illustrate that the hip and knee flex and extend together with similar magnitudes. Also, similar patterns of flexion then extension were observed for the torso and ankle during the traditional squat. Assessment of the joint angle-time curves for the powerlifting squat and box squat revealed subjects selected 1 of 2 distinct techniques to perform the movement (Figures 3 and 4 illustrate representative curves for the distinct patterns used in the powerlifting squat). The first technique exhibited similar flexion and extension angles for the hip and knee as observed during the traditional squat (Figure 3). However, the movement also included substantially more rotation of the femur around the vertical and anterior-posterior axes than observed during the traditional squat. The second technique observed exhibited 2 distinct phases during the eccentric portion of the movement (Figure 4). Initially, movement was isolated in the sagittal plane at the hip joint. Upon reaching a critical hip flexion angle, the knee and ankle simultaneously flexed along with concurrent abduction and internal rotation of the femur. Although the same overall movement patterns were observed for the powerlifting squat and box squat, the actual magnitude of torso inclination and ankle flexion during the eccentric phase were reduced when the box was introduced.

Figure 2
Figure 2:
Representative joint angle-time curve for the traditional squat. The dashed line indicates transition from eccentric to concentric.
Figure 3
Figure 3:
Representative joint angle-time curve for a distinct movement pattern observed during the powerlifting squat. The dashed line indicates transition from eccentric to concentric.
Figure 4
Figure 4:
Representative joint angle-time curve for a second distinct movement pattern observed during the powerlifting squat. The dashed line indicates transition from eccentric to concentric.

Angular Kinetics

Peak joint moments and moment arms are shown in Table 3. Moment arms were calculated relative to the barbell center and correspond with the time interval of the peak joint moment. Positive values indicate the barbell was anterior to the joint center and negative values indicate a posterior barbell location. Significant differences were obtained for all joint moments and moment arms across the exercises. The greatest differences in peak joint moments were recorded at the spine and ankle. At both joints, the largest peak moments were produced during the traditional squat, followed by the powerlifting squat, then box squat. The addition of a box resulted in significant changes to a number of moment arms and peak joint moments. In particular, the use of a box decreased peak extension moments at the spine and hip and increased peak extension moments at the knee.

Table 3
Table 3:
Peak joint moments and corresponding moment arms (mean ± SD).

External Kinematics and Kinetics

The external stimulus of each exercise was assessed through measurement of the GRF, velocity, power, and RFD. The vertical GRF maintained an overall similar profile for each exercise across loads. However, it was observed that as the external load increased the vertical GRF-time curve became more bimodal, with an increase in the relative size of the second peak. The group average vertical GRF-time curves performed with a load of 70% 1RM are shown in Figure 5. The greatest differences in vertical GRF were observed during the box squat. There were no sharp increases in force production during the transition between eccentric and concentric phases as was evident with the other exercises. In addition, as the individual sat and paused, there was a gradual transfer of load from the system to the box resulting in a substantial reduction in force production. Across the loading conditions, significantly greater peak vertical GRF was obtained for the traditional squat and powerlifting squat compared with the box squat (Table 4). Significant differences were also obtained for peak velocity, peak power, and RFD. The greatest differences were obtained for RFD where threefold to fourfold larger values were obtained for the box squat.

Figure 5
Figure 5:
Group average force-time curves obtained with a 70% 1-repetition maximum load. Circles indicate transition between phases of the squat (eccentric/concentric) and (eccentric/box/concentric).
Table 4
Table 4:
External kinematics and kinetics (mean ± SD).


The results of this study reveal significant biomechanical differences between the traditional squat and 2 of its most popular variations, the powerlifting squat and box squat. One of the most significant technical differences noted was the stance width used for each exercise. All the athletes in this study self-selected a narrow stance for the traditional squat and a wide stance for the powerlifting squat and box squat. Previous research investigating the effects of stance width on squatting biomechanics has reported a number of findings similar to the results obtained here (10). Using video data collected during a powerlifting competition, Escamilla et al. (10) reported that athletes performing wide stance squats exhibited greater hip flexion and smaller plantarflexion angles than those performing narrow stance squats. These results correspond with the significant differences in joint angles recorded in this study between the narrow stance traditional squat and the wide stance powerlifting squat. In addition, Escamilla et al. (10) reported similar effects of stance width on hip and ankle moments. In particular, wide stance squats were found to produce significantly larger hip extension moments and smaller ankle extension moments (10). In contrast to the findings of this study, Escamilla et al. (10) reported that overall joint-time curves for the torso and lower body were similar between narrow and wide stance squats. However, data collected by Escamilla et al. (10) were recorded during an active competition, and the authors were unable to influence the lifting techniques employed, whereas, in this study athletes were instructed to let the knee travel past the toes during the traditional squat and to maximize posterior displacement of the hip during the powerlifting squat and box squat. These instructions resulted in different movement strategies beyond alterations to stance width. The joint-time curves for the traditional (narrow stance) squat were consistent across subjects and featured simultaneous flexion then extension of the hip and knee, with greater range of motion obtained at the knee joint (Figure 2). During the powerlifting squat and box squat (wide stance), 2 distinct techniques were observed. The first technique also featured simultaneous flexion then extension of the hip and knee. However, the movement was combined with significantly greater abduction and adduction and internal and external rotations of the femur compared with that measured during the traditional squat (Figure 3). The second technique observed during wide stance squats featured 2 distinct phases during the eccentric portion of the movement (Figure 4). The first phase consisted of isolated hip flexion to approximately 40°. Upon reaching this point, the second phase of the movement was initiated and comprised rapid flexion of the knee and ankle, combined with substantial abduction and internal rotation of the femur. The different movement strategies selected were clearly influenced by the stance width adopted. When attempting to displace the knees past the toes, a narrow stance may have been selected to facilitate tracking of the patella over large knee flexion angles. In contrast, a wide stance was most likely adopted when attempting to maximize posterior displacement of the hip to decrease the height of the system COM and increase overall stability.

When discussing the advantages and potential risks associated with each type of squat, researchers and practitioners have generally focused on the kinetics associated with the exercise (6). Based largely on research conducted by Fry et al. (12), it is commonly believed that squats which minimize anterior displacement of the knee produce greater muscular forces at the hip and require a more horizontal torso position to remain balanced. Importantly, it is believed that this torso position results in larger forces and moments experienced at the lumbar spine, which increases the risk of developing lower back injuries. The results from this study support claims that greater muscular forces are generated at the hip when attempting to maintain a more vertical shin position (6). This conclusion is based on significant differences in peak joint moments measured between the traditional squat and powerlifting squat. In contrast to the findings of Fry et al. (12), the results obtained here demonstrate that positioning of the torso is not dependent on the amount of anterior knee displacement. In addition, the largest peak moments at the L5/S1 joint in this study were measured during performance of the traditional squat and not the powerlifting squat as would have previously been expected. Collectively, the results contradict previous suggestions that there is a greater risk of developing lower back injuries when performing variations such as the powerlifting squat. Contrasting results may be because of a number of methodological differences between the studies. The subjects recruited by Fry et al. (12) were recreationally trained and attempted to adopt similar movement strategies when performing the traditional squatting technique and the variation with restricted anterior knee displacement. Conversely, subjects in this study were competitive powerlifters with enough experience in both exercises to select different movement strategies. Based on consistent technical features adopted by all the athletes in this study, it is clear that maintaining a relatively upright torso position while restricting anterior displacement of the knee is best achieved by adopting a wide stance and achieving significant range of motion at the hip joint in all 3 planes of motion. This may have implications for individuals who wish to perform the powerlifting squat or restrict anterior displacement of the knee but have limited movement capabilities at the hip joint.

Differences in peak joint torques recorded for each exercise were largely a result of the relative displacements of the barbell and joint centers. Performance of the traditional squat created relatively large anterior displacements of the barbell, knee and system COM during the eccentric phase (Table 1). In contrast, the use of the box enabled individuals to maximize posterior displacement of the hip which resulted in an overall posterior displacement of the barbell. Visual observation of box squat repetitions revealed that many of the powerlifters displaced the system COM behind the base of support during the final stages of the eccentric movement. The use of the box to safely maximize posterior displacement created an ordered succession of squatting motions with the traditional squat situated at one end of the spectrum and the box squat at the other. A number of peak joint moments analyzed in this study reflected this ordered succession. At the ankle joint, peak extension moments were greatest during the traditional squat, followed by the powerlifting squat, then box squat. Differences in peak moments measured at the ankle would have been caused by variation in the displacement of the system COM. The larger anterior displacements created during the traditional squat would have resulted in an increased joint moment to compensate for the greater total resistance (28). Based on the results of previous research (12,28) and large differences noted across techniques for anterior knee displacement, a similar ordered effect was expected for peak moments developed at the knee joint. However, the results showed that the largest peak moments were obtained during the box squat, with similar smaller values obtained during the traditional squat and powerlifting squat. For each exercise, the peak knee extension moment was developed during the initial stage of the concentric movement. As individuals maintained a more upright torso position when performing the box squat, the greater resistance moment arm created explains the larger peak moment recorded. The magnitude of the resistance moment arm created at the knee joint was similar between the traditional squat and powerlifting squat. As a result, no significant difference for the peak knee extension moment was measured between the 2 exercises. This result contradicts findings from previous research reporting reduced knee moments when maintaining a more vertical shin position (12). However, previous results were associated with an increased forward lean of the torso, which did not occur in this study. It is also important to note that the overall mechanical stress experienced at the knee may not be adequately described by the peak moment alone. Research has shown that compressive and shear forces at the knee increase with larger flexion angles and greater displacement of the femur relative to the tibia (11,24,28). As a result, it is expected that greater overall stress at the knee joint will occur during the traditional squat.

Significant kinetic differences were also obtained at the hip joint. Across exercises, the largest peak moment was obtained during performance of the powerlifting squat. This result may be because of a number of biomechanical and physiological factors. The increased forward lean of the torso during the powerlifting squat in comparison with the box squat would have created a larger resistance moment arm at the hip, which would explain the difference in peak extension moment found. However, a significant difference was also obtained between the powerlifting squat and traditional squat despite both exercises creating a similar resistance moment arm. The difference may have been caused by variation in recruitment of the muscles surrounding the hip joint. Researchers have previously commented that powerlifters intentionally emphasize hip extension when performing wide stance squats (28). Support for this claim can be found in multiple studies that have reported increased muscle activity of the gluteus maximus when squats are performed with wider stance widths (20,22). In addition to creating the largest extension moment at the hip, the powerlifting squat also produced the largest peak abduction and peak axial rotation moments. These larger kinetic values corresponded with greater frontal and transverse rotations of the femur during the powerlifting squat compared with the other exercises. Recently, there has been interest in altering the position of the femur during squatting exercises to target specific muscle groups (11,23,25). Anecdotally, it is believed that performing the squat with the hip in external rotation increases muscle activity of the quadriceps and hip abductors (25). Research conducted thus far has failed to demonstrate changes in quadriceps activity with altered rotation of the femur (11,25); however, data exist to suggest that muscle activity of the hip abductors can be influenced (23). Previous studies have attempted to control the position of the femur by fixing the orientation of the foot. However, during this study significant axial rotation was measured despite the foot remaining still. For each exercise, the movement was initiated with the foot abducted and the hip externally rotated. As the movement progressed, foot position remained fixed as the hip moved in and then out of internal rotation. Results from other kinematic studies incorporating 3D motion capture systems have reported similar results for athletes performing the squat (9,29). This observation may have implications for potential injuries at the knee joint. Research has previously shown that hip adduction combined with internal rotation of the femur during knee flexion exercises is associated with increased valgus stress and repetitive injuries such as anterior cruciate ligament strain, iliotibial band friction syndrome, and patellofemoral pain syndrome (15,18). During the bottom portion of the squat where internal rotation of the femur was at its greatest, the athletes in this study were able to maintain appropriate alignment of the femur and tibia through substantial abduction of the hip. During the powerlifting squat where internal rotation and hip flexion is maximized, untrained individuals and those with restricted movement capabilities may be unable to maintain hip abduction. This may lead to those individuals descending into an adducted and internally rotated posture which could create large stresses at the knee.

To obtain a more complete understanding of the biomechanical stimulus presented by an exercise, recent research has focused on the external kinematics and kinetics created (7,17,30). Most frequently, variables such as force, velocity, power, and RFD have been measured (1). The data obtained have also been used to rank exercises based on the belief that those which acutely maximize the production of each variable provide the best stimulus for longitudinal improvement. To ensure the biomechanical stimulus is maximized for each variable, repetitions in the present and previous studies were performed with the intention to lift the load as fast as possible (7,17,30). The results obtained here demonstrate that large forces can be produced in all 3 squatting exercises even when light resistances are displaced with maximum velocity. Across the 30–70% 1RM loads, peak vertical GRF for the group was approximately 2.1–2.8 times body weight. The largest effects of squat variation on force and all other external kinematics and kinetics recorded were obtained during the box squat. Group average force-time curves showed reduced peak values and changes to the overall profile with the box squat compared with the other exercises (Figure 5). During the traditional squat and powerlifting squat, a large increase in force was measured during the transition period between eccentric and concentric phases. However, during the box squat, the athletes were able to decrease force production during this transition period and use the box to partially slow the system COM. After a sustained reduction in force as the athletes paused on the box, force was then rapidly increased during the concentric phase. A similar reduction in peak force when performing the box squat was reported in a recent study conducted by McBride et al. (19) The authors suggested that lower forces produced during the box squat compared with a standard squatting movement was the result of reduced stretch-shortening activity from pausing on the box. The powerlifters in this study were instructed to follow their individual practices regarding the length of time paused on the box, as long as a minimum period of 1 second was adhered to. On average, the group paused for 1.7 seconds with times ranging from 1.3 to 2.3 seconds. Research has shown that as duration between eccentric and concentric phases increases there is a progressive reduction in contribution from the stretch-shortening cycle (27). The long pauses obtained during the box squat are therefore likely to explain the reductions in force, velocity, and power in comparison to the other exercises studied. However, the largest effect of squat variation observed was an increase in RFD during the box squat. The results showed threefold to fourfold greater values in RFD when squats were performed with the box. Because RFD and the squat exercise are both considered important elements of training for athletic improvement (8), the finding that significantly larger RFD values can be obtained when using the box could have important implications for training prescription. Although it remains unclear which training practices are most effective for long-term improvements in RFD, many believe that performing explosive resistance exercises that create high RFD values will be successful (8). The large disparity in RFD values obtained between the exercises may provide researchers with an effective model to study RFD using movements that are transferable to many sporting actions.

Practical Applications

The squat is widely regarded as one of the most effective exercises for improving strength and athletic performance. Most often, the exercise is performed with a narrow stance, and the knee is permitted to travel past the toes. In many instances, strength and conditioning coaches will attempt to manipulate the exercise to target particular areas of the body or simply to provide variation in training. Traditionally, strength and conditioning coaches have manipulated the biomechanics of the squat by altering the position of the barbell to perform either the front squat or overhead squat. However, the results of this study show that the biomechanical stimulus of the squat can be altered by employing different movement strategies and by using a box to modify the transition between eccentric and concentric phases. By instructing individuals to maximize posterior displacement of the hip as is required during the powerlifting squat, it is possible to increase the stress placed on the hip joint in all 3 planes. This squatting style requires a wide stance to remain stable and if performed correctly may decrease the stress placed at the ankle and lumbar region in comparison to the traditional squat. In addition, performance of the powerlifting squat may be beneficial for individuals who have sufficient movement capabilities at the hip but lack range of motion at the ankle joint and therefore are unable to descend to sufficient depth with appropriate body positioning. Coaches and athletes should be aware that correct performance of the powerlifting squat may require substantial mobility at the hip joint and practice to coordinate the segments of the body. We recommend that the box squat be used as a training tool to improve competency in performing the powerlifting squat. Initially, a relatively tall box may be used to teach the exercise and progressed by gradually decreasing the height as proficiency increases. In addition, the very large RFD values produced during the box squat suggest it could be an effective exercise to develop explosive strength and athleticism. Based on current paradigms used in the training of athletes, it is recommended that multiple sets of 3–6 repetitions be performed to develop these qualities.


1. American College of Sports Medicine. Progression models in resistance training for healthy adults. Med Sci Sports Exerc 41: 687–708, 2009.
2. Baechle TR, Earle RW. Essentials of Strength Training and Conditioning. Champaign, IL: Human Kinetics, 2008.
3. Baumgartner TA. Reliability and error of measurement. In: Measurement Theory in Practice and Kinesiology. T. M Wood, W Zhu, eds. Champaign, IL: Human Kinetics, 2006. pp. 27–52.
4. Brown LE, Shepard G, Sjostrom T. Point/counterpoint: Performance box squats. Strength Cond J 25: 22–23, 2003.
5. Chandler TJ, Stone MH. The squat exercise in athletic conditioning: A position statement and review of the literature. Strength Cond J 13: 51–60, 1991.
6. Chiu LZF, Heiler J, Sorenson SC. Sitting back in the squat. Strength Cond J 31: 25–27, 2009.
7. Cormie P, McCaulley GO, Triplett TN, McBride JM. Optimal loading for maximal power output during lower-body resistance exercises. Med Sci Sports Exerc 39: 340–349, 2007.
8. Cronin J, Sleivert G. Challenges in understanding the influence of maximal power training on improving athletic performance. Sports Med 35: 213–234, 2005.
9. Decker M, Krong J, Peterson D, Anstett T, Torry M, Giphart E, Shelburne K, Philippon M. Deep hip muscle activation during a squat exercise. In: Proceedings of the American Society of Biomechanics. PA: 2009.
10. Escamilla RF, Fleisig GS, Lowry TM, Barrentine SW, Andrews JR. A three-dimensional biomechanical analysis of the squat during varying stance widths. Med Sci Sports Exerc 33: 984–998, 2001.
11. Escamilla RF, Fleisig GS, Zheng N, Lander JE, Barrentine SW, Andrews JR, Bergemann BW, Moorman CT. Effects of technique variations on knee biomechanics during the squat and leg press. Med Sci Sports Exerc 33: 1552–1566, 2001.
12. Fry AC, Smith C, Schilling BK. Effect of knee position on hip and knee torques during the barbell squat. J Strength Cond Res 17: 629–633, 2003.
13. Gutierrez-Farewik EM, Bartonek A, Saraste H. Comparison and evaluation of two common methods to measure center of mass displacement in three dimensions during gait. Hum Mov Sci 25: 238–256, 2006.
14. Hales ME, Johnson BF, Johnson JT. Kinematic analysis of the powerlifting style squat and the conventional deadlift during competition: Is there a cross-over effect between lifts? J Strength Cond Res 23: 2574–2580, 2009.
15. Ireland ML. The female ACL: Why is it more prone to injury? Orthop Clin North Am 33: 637–651, 2002.
16. Kadaba MP, Ramakrishnan HK, Wooten ME. Measurement of lower extremity kinematics during level walking. J Orthop Res 8: 383–392, 1990.
17. Kawamori N, Rossi SJ, Justice BD, Haff EE, Pistilli EE, O'Bryant HS, Stone MH, Haff GG. Peak force and rate of force development during isometric and dynamic mid-thigh clean pulls performed at various intensities. J Strength Cond Res 20: 483–491, 2006.
18. Leetun DT, Ireland ML, Willson JD, Ballantyne BT, Davis IM. Core stability measures as risk factors for lower extremity injury in athletes. Med Sci Sports Exerc 36: 926–934, 2004.
19. McBride JM, Skinner JW, Schafer PC, Haines TL, Kirby TJ. Comparison of kinetic variables and muscle activity during a squat vs. a box squat. J Strength Cond Res 24: 3195–3199, 2010.
20. McCaw ST, Melrose DR. Stance width and bar load effects on leg muscle activity during the parallel squat. Med Sci Sports Exerc 31: 428–436, 1999.
21. McLaughlin TM, Dillman CJ, Lardner TJ. A kinematic model of performance in the parallel squat by champion powerlifters. Med Sci Sports Exerc 9: 128–133, 1977.
22. Paoli A, Marcolin G, Petrone N. The effect of stance width on the electromyographical activity of eight superficial thigh during back squat with different loads. J Strength Cond Res 23: 246–250, 2009.
23. Pereira GR, Leporace G, Chagas DV, Furtado LFL, Praxedes J, Batista LA. Influence of hip external rotation on hip adductor and rectus femoris myoelectric activity during a dynamic parallel squat. J Strength Cond Res 24: 2749–2754, 2010.
24. Schoenfeld BJ. Squatting kinematics and kinetics and their application to exercise performance. J Strength Cond Res 24: 3497–3506, 2010.
25. Signorlie JF, kwiatkowski K, Caruso JF, Robertson B. Effect of foot position on the electromyographical activity of the superficial quadriceps muscles during the parallel squat and knee extension. J Strength Cond Res 9: 182–187, 1995.
26. Swinton PA, LLoyd R, Agouris I, Stewart A. Contemporary training practices in elite British powerlifters: Survey results from an international competition. J Strength Cond Res 23: 380–384, 2009.
27. Wilson GJ, Murphy AJ, Pryor JF. Musculotendinous stiffness: its relationship to eccentric, isometric, and concentric performance. J Appl Physiol 76: 2714–2719, 1994.
28. Wretenberg P, Feng Y, Arborelius UP. High- and low-bar squatting techniques during weight-training. Med Sci Sports Exerc 28: 218–224, 1996.
29. Wu H, Lee S, Kao J, Wang S. Three-dimensional kinetic analysis of lower limbs in barbell squat. In: Proceedings of 37th Annual Northeast Bioengineering Conference. Rochester, NY: 2011.
30. Zink AJ, Perry AC, Robertson BL, Roach KE, Signorile JF. Peak power, ground reaction forces, and velocity during the squat exercise performed at different loads. J Strength Cond Res 20: 658–664, 2006.

kinematics; kinetics; RFD; submaximum loads; technique

© 2012 National Strength and Conditioning Association