The resisted squat is a poorly understood exercise across the fitness industry, particularly when squat depth is being discussed. A “deep squat” can be defined as a squat where the hips are well below the horizontal plane of the knees or, alternatively, it may be defined as the hips reaching the same horizontal plane as the knees. If the former definition (i.e., hips below knees) is used for a deep squat, then the traditional “full squat” (i.e., hips parallel to knees) is better categorized as a “partial squat.” However, if the latter definition of a full squat is used (hips parallel to knees), then a partial squat may be considered a squat where the knees are flexed to only 90–120° (6). Clearly, a lack of standardized terminology has led to a great deal of confusion about the appropriate depth of squatting.
Although research from Klein (25) found high shearing force through the knee during “full squats,” the squats investigated were squats in which the hips were below the knees. As a result of parallel squats being confused with full squats, parallel squats have been criticized unfairly as being of a high injury risk (32). With deep squats often listed as a contraindicated exercise (28), it is not unusual for many coaches and trainers to believe that this contraindication applies to all squatting movements (9). No research has ever shown that parallel squats create high shearing force through the knee (18). Parallel squats are not only safe when performed correctly (10) but are an integral part of effective fitness training programs (7,8,33).
In response to the claims that partial squats are safer than full squats, partial squats have become widespread in the fitness industry (9). Limiting the range of motion of the squat may improve sport-specific adaptations for athletes by increasing the movement specificity to a sporting action. Many coaches have modified the squat range of motion from parallel (i.e., hips parallel to knees) to partial squats (e.g., 120° of knee flexion) to incorporate movements that more closely resemble the lower range of motion (36). Given the technical difficulty in coaching and performing parallel squats, partial squats are also regularly taught and performed in community-based gymnasiums. Limited knowledge of many trainers and trainees and poor instruction in the squat exercise is an issue (11).
Caterisano et al. (6) concluded that, of the hip extensor muscles, only gluteus maximus activation increased as squat depth increased. However, Jensen and Ebben (23) concluded that only hamstring activation increases with increasing squat depth and only in the top portion of the squat. Although several studies have quantified shearing force though the knee (18) and neuromuscular properties (6,23) at different squat depths, none have compared movement kinetics. Few training studies have used partial squats training in an effort to increase sporting performance measures such as sprinting and jumping (36), let alone comparing PROM with FROM resistance training programs (35). At the conclusion of a 10-week PROM squat training protocol, Wilson et al. (36) reported an increase of 7.1% in countermovement jump height and 4.9% in static jump height when training squat depth to knee angles of between 120 and 180° at moderate to heavy loads. Weiss et al. (35) found no significant difference between training FROM squats and PROM squats over a 6-week training program on vertical jump and depth jump ability in untrained individuals. Therefore, the influence of squat depth as a variable in squat training remains unclear.
Although electromyography studies generally demonstrate greater muscle activation during full squatting, it is not clear if there are any functional (kinetic) differences. If squatting to parallel is safe and produces superior movement kinetics, there is a rationale for supporting parallel depth squatting. Therefore, the purpose of this study is to directly compare the kinetic properties of FROM and PROM squats at different intensities.
Experimental Approach to the Problem
We sought to compare changes in movement kinetics when squatting with the same intensity with different depths and to compare changes in movement kinetics when squatting with different intensities with the same depth. The participants performed 1 of 4 randomly assigned squat protocols per day over a 14-day period with each session separated by at least 3 days. Independent variables manipulated in each squat protocol included either the squat intensity (67% 1 repetition maximum [1RM] or 83% 1RM) or the depth (knees to 120° or hips parallel to knees) of the squat performed. Dependent variables collected during each squat protocol via optical encoder included movement distance, concentric peak velocity, concentric peak power, concentric peak force, and total concentric work.
Ten male recreational team-sport athletes who played in a locally based rugby competition volunteered for the study. Participants' training comprised regular team practice twice per week, and they were involved in resistance training 2–3 times per week. Each participant had at least 1 year of squat training experience. The participants had a mean (±SD) age of 21.4 (±1.14) years, height 180.3 (±6.5) cm, body mass 88.5 (±15.6) kg, 1RM FROM 148.83 (±28.75) kg, and 1RM PROM 270.78 (±51.7) kg. Before commencing the study, the participants were screened for a recent history of injuries (e.g., knee, lower back) that may influence their performance in the study. There were no positive responses to verbal enquiries regarding illegal performance enhancing substances. All testing sessions occurred at a similar time of day for each participant, and the participants were asked to maintain a consistent diet, including fluids and caffeine consumption in the 24 hours before testing. The experiment was approved by the Human Research Ethics Committee of the university, and written informed consent was obtained following an information presentation that provided full explanation of the procedures involved.
The 2 independent variables manipulated in this experiment were squat depth (PROM or FROM) and squat intensity (67% 1RM or 83% 1RM). For PROM squats, the hips and knees were flexed until the participant reached 120° of knee extension, whereas for FROM squats, the hips and knees were flexed until the hips reached the same horizontal plane as the knees. To determine the desired knee angle, each participant was asked to use an unloaded 20-kg bar to squat to the desired position. The angle of knee flexion was confirmed using a manual goniometer for PROM squats. A tightly drawn rubber cord was secured behind the participant's legs touching the back of the thighs to mark this goal depth. The rubber cord was used to ensure compliance to the goal depth as the participant contacted the cord with the thighs during each repetition. The rubber cord was tied in such a way that the knot would slip if the participant failed to lift the weight and collapsed under the weight. Safety bars were adjusted to a depth that would safely support the barbell if the participant failed to lift the weight. Several familiarization trials were completed with the 20-kg bar to ensure the participants returned to the correct position.
Technical criteria for both squat techniques involved cradling the bar on the back in the “high bar” position with feet placed approximately shoulder width apart, and the toes pointed slightly outward based on the preference of the participant. Feet and heels remained flat on the floor and the back flat at all times (31). The knees and hips were flexed to the goal depth and then fully extended to the beginning position. The bar was not allowed to stop at anytime during sets (17).
The participants' 1RM for both FROM and PROM squats were collected 3 days before the commencement of the study. After a warm-up of 2 sets of FROM squats that did not cause failure, a weight that could be squatted not >15 times before failure (2) using a PROM was loaded on the 20-kg bar. After a 20-minute rest period, the participant repeated the previous warm-up. A weight that could be squatted not >15 times before failure using a FROM was loaded on the 20-kg bar. The participants' FROM 67%1RM, FROM 83%1RM, PROM 67%1RM, and PROM 83%1RM were calculated using a prediction equation based on the resistance and number of repetitions completed (2).
Subsequent testing sessions involved the participant performing 4 sets of either FROM squats for 10 repetitions (FROM10) at an intensity of 67%1RM FROM squat, PROM squats for 10 repetitions (PROM10) at 67%1RM PROM squat, FROM squats for 5 repetitions (FROM5) at 83%1RM FROM squat, or PROM squats for 5 repetitions (PROM5) at 83%1RM PROM squat (Table 1). During all testing sessions, there was a 90-second rest interval between sets. The speed of the squat was performed at and unintentional velocity (i.e., at the participant's own discretion) (16,22).
Dependent variables collected during each squat protocol via optical encoder included movement distance, concentric peak velocity, concentric peak power, concentric peak force, and total concentric work. Kinetic data were collected using a GymAware Powertool (Kinematic Performance Technology, Canberra, Australia). The GymAware Powertool is a spring loaded retractable cable that passes around a spool integrated with an optical encoder. The external end of the cable is attached to the barbell. The GymAware Powertool is placed on the floor directly below the movement of the bar. The encoder records velocity and the movement distance of the cable. The data output is collected and stored in a personal digital assistant (Tungsten-e, Palm, Milpitas, CA, USA). The mass of the barbell is entered in the personal digital assistant and integrated with the optical encoder data to give power output for every 3 mm of bar movement.
We have previously validated the use of the GymAware Powertool for assessing kinetics of barbell movements (15). To assess reliability, squat kinetics were collected on 2 repeat trials in pilot testing (unpublished data, n = 9) to quantify the within-subject variation (typical error measurement [TEM]) from 1 trial to the next (20). The TEM for each of the squat kinetics using the optical encoder was established as power 2.8%, velocity 2.5%, work 3.9%, repetition duration 5.3%, depth 3.5%, and force <1%.
Descriptive statistics are expressed as mean and SD. Each dependent variable (i.e., movement distance, concentric peak force, concentric peak velocity, concentric peak power, and concentric work) was compared across each of the squat conditions (i.e., PROM10, PROM5, PROM10, PROM5) by repeated-measures analysis of variance (ANOVA) using the Predictive Analytics SoftWare (PASW version 17.0.2. Chicago, IL, USA). Results of the repeated-measures ANOVA are expressed as mean difference between any 2 conditions with 95% confidence limits. Magnitudes of mean differences are expressed with standardized (Cohen) effect sizes (21,37); thresholds for qualitative descriptors of Cohen's d were set at <0.20 is "trivial," 0.20–0.49 is "small," 0.50–0.79 is "moderate," and >0.80 is “large” (12). Data are expressed using confidence limits and effect sizes to express richer detail regarding the clinical applications of results (21,37).
The participants produced higher mean concentric peak power during PROM5 (1,051 ± 165 W) than FROM10 (796 ± 151 W), PROM10 (953 ± 221 W), or FROM5 (883 ± 209 W). Peak power produced during PROM5 was substantially more than the PROM10 (98 W, −21 to 217, mean, lower and upper 95% confidence limits), FROM5 (168 W, 47–28), and FROM10 (255 W, 145–365) (Figure 1). The FROM5 also generated “moderately” more power than FROM10 (87 W, 21–153, d = 0.57).
The participants produced more mean concentric peak force of during PROM5 (2,192 ± 366 N) than FROM10 (1,123 ± 195 N), PROM10 (1,820 ± 404 N), or FROM5 (1,338 ± 250 N). The force produced during PROM5 was substantially more than PROM10 (372 N, 254–490), FROM5 (854 N, 731–977), and FROM10 (1,069 N, 911–1,227) (Figure 2).
The participants produced higher peak velocity during FROM10 (0.83 ± 0.13 m·s−1) than PROM10 (0.58 ± 0.06 m·s−1), FROM5 (0.72 ± 0.13 m·s−1), or PROM5 (0.52 ± 0.08 m·s−1). The peak velocity produced during FROM10 was substantially more than FROM5 (0.105 m·s−1, 0.044–0.166), PROM10 (0.246 m·s−1, 0.167–0.325), and PROM5 (0.305 m·s−1, 0.228–0.382) (Figure 3).
The participants produced more mean concentric work during FROM5 (582 ± 67 J) than FROM10 (496 ± 57 J), PROM10 (372 ± 74 J), or PROM5 (441 ± 78 J). The total concentric work produced during FROM5 was substantially more than FROM10 (86 J, 59–113), PROM5 (142 J, 90–194), and PROM10 (211 J, 165–257) (Figure 4). Also of interest, FROM10 generated a “large” amount more total work per repetition than PROM10 (125 J, 84–166, d = 2.18).
The mean squat depth was only relevant when comparing different intensities in the same intended range of motion. The participants were able to squat deeper during the FROM10 condition (58.0 ± 6.1 cm) than the FROM5 (53.8 ± 9.5 cm) by a “moderate” 4.2-cm difference (0.4–8.0, d = 0.70). The difference between PROM10 (25.6 ± 3.9 cm) and PROM5 (25.0 ± 3.9 cm) was a “trivial” 0.60 cm (−2.0 to 3.2, d = 0.13).
The purpose of this research was to assess the influence of squat depth at different relative intensities on kinetics of squatting. The results show that power and force were greatest when squatting in the partial range of motion (PROM) at the heaviest load, whereas movement velocity was highest in the full range of motion (FROM) at the lightest load. Additionally, FROM squats resulted in a higher amount of total concentric work being performed per repetition when compared with their PROM counterparts; work output was highest in the FROM with the heavier load. Effect sizes for all of these differences were at least “moderate” and typically “large.” Also of interest was that squatting at the 10RM load in the PROM was not highest in any of force, power, velocity, or work when compared with other ranges of motion or intensities. Therefore, all of PROM5, FROM5, and FROM10 can be effective training modalities depending on the training goal; however, squatting in the PROM at 75%1RM (i.e., 10RM) is not optimal for developing any movement kinetic.
Our results demonstrate that the participants produced greater power when using heavier loads than lighter loads when performed in the same range of motion. This outcome is somewhat surprising because, 30–60% 1RM is recommended to maximize the interaction between force and velocity to produce the highest peak power output when the participant is purposefully attempting to move at maximal velocity (30). However, we observed that, when using unintentional movement velocity, the heaviest load produced the greatest peak power. An earlier study supports our finding that the greatest power output was achieved using the greatest resistance without the use of conscious velocity control (29). In this study, although the velocity was 15% higher in FROM10 than in FROM5 and 11% higher in PROM10 than in PROM5, force was 19 and 20% higher, respectively, in both ranges of motion at the heavier load. This combination resulted in power being approximately 10% higher in both ranges of motion at the heavier load in the same range of motion. Therefore, although we observed the lowest peak movement velocity at the heaviest load (i.e., FROM10 was 59% faster than PROM5), force increased disproportionately more (PROM5 produced 95% more force than FROM10) thereby resulting in 32% higher peak power for PROM5 than for FROM10.
The force-power relationship (13) indicates that power output peaks when force is approximately 30% of maximal. The force-power relationship presumes that the participant is intending to move at maximal velocity, however. The unintentional movement velocity in this study is an important consideration when explaining how peak power resulted in the higher loads: The participants were likely moving at a much higher proportion of their maximal voluntarily velocity during the heavier squats while during the lighter loads they had a tendency to “cruise” because of the lower intensity of the resistance. Without conscious control of movement velocity, training heavier loads likely inadvertently results in higher power output. Whether the same outcome would occur if maximal velocity was emphasized on every concentric movement is unknown though seems likely based on previous literature (13). Heavy-load training has yielded improvements in laboratory as well as field-based tests when the intent to move at maximal velocity is emphasized, even if the actual movement is quite slow (3,24). Other research has shown that heavy-load resistance training can increase power (1) and athletic performance (4,27) even without “maximal intent” of velocity. Therefore, the low movement velocities experienced in high-load resistance training may still be capable of improving sport-specific movements (4,27,36) and warrant consideration in strength training program design if a specified range of motion is not an important consideration.
Although power peaked at the heaviest load when training in the PROM, velocity was maximized at the lightest load in the FROM. These findings reflect Hills' (19) classic relationship between load and velocity. This finding is of particular relevance to athletes where performance requires high velocity. We suggest that training FROM10 should be incorporated into training routines that require high-velocity demands such as sprinting- or jumping-based athletes.
Although many sporting events require high-velocity demands for competitive success, others require force. During the PROM5 condition, the participants lifted the heaviest load, whereas during the FROM10 condition the participants lifted the lightest load. Where force output rather than a high range of motion is a key determining factor to sporting performance (e.g., rugby scrummaging), training PROM5 should be considered. Other sports require both high velocity and high force production to be successful; velocity and force may both be important factors in their sporting success (14,27). Therefore, rather than considering squats as a single exercise, depending on the goal of the training session, it may be beneficial for athletes to train using both partial and full squats. For example, FROM10 squats could be used to target the velocity-specific components of a sport and, in a separate training session or training phase, PROM5 trained to target the high force and high power components of the sport (5).
There were also substantial differences in squat depth and total concentric work per repetition. The FROM5 squats were a “moderate” approximately 5 cm less deep than the FROM10 squats, whereas the difference between the 2 PROM intensities was “trivial.” That the FROM10 squats were substantially deeper than the FROM5 squats was most likely a consequence of the participants being reluctant to descend further than a depth they could confidently squat given the increased difficulty of executing the concentric movement with a heavier load (16). When purposefully training the FROM, the participants should be monitored carefully for squat depth as load increases. We observed that squatting with FROM10 and FROM5 produced 34 and 32% greater work output per repetition in comparison to PROM10 and PROM5 squats, respectively, whereas squatting with FROM5 produced 17% more work than FROM10. Differences in work output were likely a consequence of differences in squat depth and weight lifted. Given the importance of work performed in the regulation of anabolic hormones (26), training program designs that maximize work output should be used for changing body composition. We suggest that performing FROM5 sets may be more beneficial when work output and time under tension is desired (e.g., fat loss or hypertrophy programs) than FROM10, even though with half as many repetitions per set twice as many sets would need to be performed to equate for the total number of repetitions.
In conclusion, training with heavy PROM squats elicited the greatest force and power, whereas training with heavy FROM squats produced the greatest work per repetition. We observed that training with moderately loaded FROM squats produced the highest peak velocity. Finally, training with moderately loaded PROM squats did not excel in any kinetic variable studied here. As a result, we recommend that a variety of squat depths and loads should be prescribed depending on the individual athlete's muscle characteristics and training goals.
The participants involved in resistance training should consider the use of near-maximal PROM squats if their goal is to exert maximal force or power through a limited range of motion. Because resistance training programs designed to change body composition (e.g., hypertrophy, fat loss) are reliant not so much on power and force but more on total work performed, high-intensity FROM squats should be the focus. Because half as many repetitions are as the FROM10 condition, twice as many sets will be needed to ensure a sufficient total work is performed. Because the highest velocity was achieved in the FROM10 condition, athlete seeking goals dependent on high velocity, such as running and jumping, should employ predominantly lighter, FROM squats. The set design that seemed to have no discernable advantages was PROM10 because it did not maximize work, velocity, force, or power. Unfortunately, most recreationally training individuals are aware of the necessity for high volume training for hypertrophy from bodybuilding magazines (34) and thus target the 8–12 repetition range with a moderate load (30). Recreational athletes also tend to have low skill in performing full depth squats and pay little conscious attention to movement velocity. Partial repetition squats performed in the moderate to high repetition range seems to have the least benefit.
1. Adams K, Oshea JP, Oshea KL, Climstein M. The effect of six weeks of squat, plyometric and squat-plyometric training on power production. J Appl Sport Sci Res 6: 36–41, 1992.
2. Baechle TR, Earle RW, Wathen D. Resistance training
. In: Essentials of Strength Training and Conditioning. Baechle T. R, Earle R. W, eds. Champaign, IL: Human Kinetics, 2008. pp. 381–412.
3. Behm DG, Sale DG. Intended rather than actual movement velocity determines velocity-specific training response. J Appl Physiol 74: 359–368, 1993.
4. Blazevich A, Jenkins D. Physical performance differences between weight-trained sprinters and weight trainers. J Sci Med Sport 1: 12–21, 1998.
5. Bompa T. Periodization—Theory and Methodology of Training. Adelaide, South Australia: Human Kinetics, 2000.
6. Caterisano A, Moss R, Pellinger T, Woodruff K, Lewis V, Booth W, Khadra T. The effect of back squat depth on the EMG activity of 4 superficial hip and thigh muscles. J Strength Cond Res 16: 428–432, 2002.
7. Chandler TJ, Stone MH. N.S.C.Aposition paper: The squat exercise in athletic conditioning: A position statement and review of the literature. Natl Strength Cond Assoc J 13: 51–58, 1991.
8. Chandler TJ, Stone MH. The squat exercise in athletic conditioning: A review of the literature. Natl Strength Cond Assoc J 13: 51–58, 1991.
9. Chandler TJ, Wilson D, Stone MH. A Survey: The squat exercise: Attitudes and practices of high school football coaches. Strength Cond J 11: 30–36, 1989.
10. Chandler TJ, Wilson GD, Stone MH. The effect of the squat exercise on knee stability. Med Sci Sports Exerc 21: 299–303, 1989.
11. Chiu L. Sitting back in the squat. Strength Cond J 31: 25–27, 2009.
12. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates, 1988.
13. Cormie P, McGuigan MR, Newton RU. Developing maximal neuromuscular power: Part 1—Biological basis of maximal power production. Sports Med 41: 17–38, 2011.
14. Cronin JB, Hansen KT. Strength and power predictors of sports speed. J Strength Cond Res 19: 349–357, 2005.
15. Drinkwater EJ, Galna B, Pyne DB, Hunt PH, McKenna MJ. Validation of an optical encoder
during free weight resistance movements and analysis of bench press sticking point power during fatigue. J Strength Cond Res 21: 510–517, 2007.
16. Drinkwater EJ, Pritchett EJ, Behm DG. Effect of instability and resistance on unintentional squat lifting kinetics. Int J Sports Physiol Perform 2: 400–413, 2007.
17. Earle RW, Baechle TR. Resistance training
and spotting techniques. In: Essentials of Strength Training and Conditioning. Baechle T. R, Earle R. W, eds. Lower Mitcham, South Australia: Human Kinetics, 2008. pp. 325–376.
18. Escamilla RF. Knee biomechanics of the dynamic squat exercise. Med Sci Sports Exerc 33: 127–141, 2001.
19. Hill AV. The heat of shortening and dynamic constants of muscle. Proc R Soc Lond B 126: 136–195, 1938.
20. Hopkins WG. Measures of reliability in sports medicine and science. Sports Med 30: 1–15, 2000.
21. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc 41: 3–13, 2009.
22. Izquierdo M, González-Badillo JJ, Häkkinen K, Ibáñez J, Kraemer WJ, Altadill A, Eslava J, Gorostiaga EM. Effect of loading on unintentional lifting velocity declines during single sets of repetitions to failure during upper and lower extremity muscle actions. Int J Sports Med 27: 718–724, 2006.
23. Jensen RL, Ebben WP. amstring electromyographic response of the back squat at different knee angles during eccentric and concentric phases. In: Presented at XVIII International Symposium of Biomechanics in Sports. Hong Kong,2000.
24. Jones K, Hunter G, Fleisig G, Escamilla R, Lemak L. The effects of compensatory acceleration on upper-body strength and power in collegiate football players. J Strength Cond Res 13: 99–105, 1999.
25. Klein KK. The deep squat exercise as utilized in weight training for athletes and its effect on the ligaments of the knee. J Assoc Phys Ment Rehabil 15: 6–11, 1961.
26. Leite RD, Prestes J, Rosa C, De Salles BF, Maior A, Miranda H, Simão R. Acute effect of resistance training
volume on hormonal responses in trained men. J Sports Med Phys Fitness 51: 322–328, 2011.
27. Liow D, Hopkins W. Velocity specificity of weight training for kayak sprint performance. Med Sci Sports Exerc 35: 1232–1237, 2003.
28. Navy and Marine Corps Public Health Center. Navy physical training series: Contraindicated exercises. Norfolk, VA, 2008.
29. Newton RU, Kraemer WJ. Developing explosive muscular power: Implications for a mixed methods training strategy. Strength Cond 16: 20–31, 1994.
30. Ratamess NA, Alvar BA, Evetoch TK, Housh TJ, Kibler BW, Kraemer WJ, Triplett NT. American College of Sports Medicine position standProgression models in resistance training
for healthy adults. Med Sci Sports Exerc 41: 687–708, 2009.
31. Rippetoe M. Let's learn how to coach the squat. Strength Cond J 23: 11–12, 2001.
32. Rogers L, Sherman T. Leg press versus squat. Strength Cond J 23: 65–69, 2001.
33. Schoenfeld BJ. Squatting kinematics and kinetics and their application to exercise performance. J Strength Cond Res 24: 3497–3506, 2010.
34. Schwarzenegger A. May the force be with you, In: Joe Weider's Muscle & Fitness. Boone, IA: AMI – Weider Publications, 1999, p 190.
35. Weiss LW, Fry AC, Wood LE, Relye GE, Melton C. Comparative effects of deep versus shallow squat and leg-press training on vertical jumping ability and related factors. J Strength Cond Res 14: 241–247, 2000.
36. Wilson G, Newton R, Murphy A, Humphries B. The optimal training load for the development of dynamic athletic performance. Med Sci Sports Exerc 25: 1279–1286, 1993.
37. Ziliak ST, McCloskey DN. The cult of statistical significance, In: The Proceedings of the Joint Statistical Meetings. Washington, DC: American Statistical Association, 2009, pp 2302–2306.