Secondary Logo

Journal Logo

Original Research

Comparison of Hamstring and Gluteus Muscles Electromyographic Activity while Performing the Razor Curl vs. the Traditional Prone Hamstring Curl

Oliver, Gretchen D1; Dougherty, Christopher P2

Author Information
Journal of Strength and Conditioning Research: November 2009 - Volume 23 - Issue 8 - p 2250-2255
doi: 10.1519/JSC.0b013e3181b8d34b
  • Free



It has become common knowledge that females tear their anterior cruciate ligament (ACL) through noncontact means anywhere from 2 to 8 times greater than their male counterparts (6). However, the question still remains as to why females are more prone to the noncontact ACL injury than males. Commonly, it has been accepted that all the proposed culprits of ACL injuries, motor control issues, are the easiest to modify (3,5,8,10).

Hewett et al. has explained that females tend to be more quadriceps dominant during athletic endeavors (3,12). Quadriceps dominance means that the quadriceps muscle group is activated prior to other muscle activation when typically one would not want the quadricep muscle group being activated first. With quadricpes dominance, the quadriceps are activated first during activities where it would be more advantageous for the hamstrings to fire either simultaneously with the quadriceps or even prior to the quadriceps activation.

Biomechanically, the quadriceps muscle group exhibits an anterior shearing force on the tibia, while the hamstring complex exhibits a more posterior force on the tibia resisting the actions of the quadriceps. If the quadriceps remain dominant in such athletic endeavors as jump landings, then there will be a decreased resistance to anterior tibial stress. The decrease in anterior tibial stress is due to the pull of the quadriceps on the tibia and the lack of force produced by the hamstrings as a result of the quadriceps dominance. Typically during jump landings, one would prefer the hamstrings to be activated first in attempt to resist undue anterior tibial stress. The inability to resist anterior tibial stress automatically predisposes one to an ACL injury (7).

The hamstrings become the victim of quadriceps dominance when they are weak. If the hamstrings are weak, then they are unable to establish the appropriate firing timing during functional activities. The ability to perform proper hamstring activation during functional activities is paramount to injury prevention as well as improving athletic performance. If one is not training the musculature functionally, then one is neglecting a major component of all athletic training or even injury rehabilitation. The functionality of training is essential since the ultimate goal of any training program should be to train the athlete to perform at maximum in a functional position.

A major issue with the hamstring muscle group not only being a victim of quadriceps dominance is that they are typically trained in a prone position. Training the hamstring while prone is not considered a functional position. The hamstrings are a 2-joint muscle group. Biomechanically, if one wants to focus on the function of the hamstrings to flex the knee, then the portion of the muscle crossing the hip should be on stretch. Placing the most proximal aspect of the hamstring muscle group on stretch allows for a greater muscular advantage at the distal end of the muscle to flex the knee. In order to place the hamstrings on stretch at the hip, the hip must be placed in flexion. With the hip in flexion, the hamstring possess optimal strength at the knee. However, even if the hip is flexed and the hamstrings are weak, then the hamstrings will still be unable to contract maximally for appropriate knee flexion in such activities as jump landings. In attempt to train the hamstrings to their fullest functional capacity, it has been shown that the razor curl training technique allows for maximum activation of the hamstring and gluteals muscle groups (11). The razor curl has been deemed effective of hamstring and gluteals muscle activation while placing the hip and knee at 90° flexion, also known as the athletic position (11). The razor curl has the total body extended and then requires the hips and knees to flex to 90° simultaneously with full contraction of the hamstrings to further the knee flexion. The net effect of training in this position is to perform an eccentric hamstring contraction both in flexion and extension in the functional athletic position.

Traditionally, one has trained the hamstring in a prone position, known as the prone hamstring curl. It was the purpose of this study to compare muscle activation of the medial hamstring and biceps femoris as well as the gluteus maximus and medius muscles during the razor curl to that of the muscle activation during the traditional prone hamstring curl. It was hypothesized that the 2 exercises would not be significantly different in their muscle activation, thus, showing the effectiveness of the razor curl as compared to the traditional prone curl.


Experimental Approach to the Problem

Descriptive statistics of means and SDs of muscle activations of the medial hamstring, lateral hamstring, gluteus maximus, and gluteus minimus were compared to determine if training in a functional athletic position is as effective if not more effective than training in a prone position by examining the normalized electromyographic (EMG) data as the average percent of their maximum voluntary isometric contraction (MVIC). A paired t-test was used in comparing each muscle's peak activation during performance of the razor curl and the traditional prone curl. The independent variable was exercise and the dependent variable was peak muscle activation. All assumptions for a paired t-test were met including independence of observations, random sampling, and homogeneity of variance (2).


Eight healthy, female intercollegiate athletes (mean age 20.8 ± 3.9 y; mean height, 177.8 ± 10.9 cm; mean weight, 67.3 ± 9.9 kg), training at the Division I level, participated in the study during their off season. The subjects all were involved in resistance training at the time of data collection. Before participation, subjects were informed of all possible risks and signed a consent form approved by the University of Arkansas Institutional Review Board.


The EMG data were collected using 3M Red-Dot bipolar surface electrodes, placed over 4 muscle bellies on the subject's dominant side according to the method of Basmajian and Deluca with an interelectrode distance of 25 mm (1,4). The muscles targeted were the following: medial hamstring (semimembranosus and semitendinosus), biceps femoris, gluteus medius, and gluteus maximus. Surface electrodes were chosen because they were noninvasive and were able to reliably detect surface muscle activity.

Prior to electrode placement, subject's skin was shaved, abraded, and cleaned with alcohol. Adhesive 3M Red-Dot electrodes were placed over the muscle bellies and parallel to the direction of the underlying muscle fibers. To assure proper electrode placement, manual muscle tests were performed through MVICs based on the work of Kendall et al. (9). Three manual muscle tests were performed, by a certified athletic trainer for a total of 5 s for each muscle group. The first and last second of each MVIC trails were removed from the data in attempt to obtain steady-state results for each of the muscle groups. The manual muscle testing provided a baseline reading for which all EMG data were based.


Following electrode placement, subjects were explained the protocol of the functional hamstring exercise, the razor curl, as well as viewed repeated demonstrations. In addition to the razor curl, all subjects also reviewed the protocol of the typical prone hamstring curl. Each subject performed several warm-up trials with verbal feedback on proper technique prior to any recording. Once each subject performed the exercise properly for consecutive trials, the recording began.

After electrodes were placed on the skin and manual muscle testing was complete, each subject performed 5 repetitions of the razor curl exercise. The razor curl has the total body extended and then requires the hips and knees to flex to 90° simultaneously with full contraction of the hamstrings to further the knee flexion (Figures 1-3). There was no time allotment of rest in-between sets. During the trials, subjects were instructed on proper posture through verbal cues. In addition to EMG data, video data were also collected from a 90° lateral view to assure appropriate technique as well as to event mark trials. All trials were event marked for pull and push phases.

Figure 1
Figure 1:
Starting position of the razor curl.
Figure 2
Figure 2:
Half-way point where the knees and hips are flexed to 90°.
Figure 3
Figure 3:
End position; hips should maintain 90° flexion and knees should be maximally flexed by full contraction of the hamstring.

After subjects completed the trials of the razor curl, they also performed 5 trials of the traditional prone hamstring curl. The traditional prone hamstring curl positions the body prone with the hips slightly flexed to about 20-30°. The subject then flexed at the knees to activate the hamstring muscle group (Figures 4 and 5). There was no time allotment of rest in-between sets. During the trials, subjects were instructed on proper posture through verbal cues. In addition to EMG data, video data were also collected from a 90° lateral view to assure appropriate technique as well as to event mark trials. All trials were event marked for concentric and eccentric phases.

Figure 4
Figure 4:
Starting position of the prone curl.
Figure 5
Figure 5:
Ending position of the prone curl.

Electromyographic Analysis

A Myopac Jr 10 channel amplifier (RUN Technologies Scientific Systems, Laguna Hills, CA, USA) transmitted the all EMG raw data at 60 Hz via a fiber optic cable to the receiver unit. The EMG unit has a common mode rejection ratio of 90 dB. The gain for the surface electrodes was set at 2000. The EMG data were recorded, stored, and analyzed with the analog data acquisition package of Peak Motus Software (version 9.0; Peak Performance, Englewood, CO, USA).

The EMG enveloped data were assessed. Mean maximum EMG reference values were calculated for each muscle within the phase. Five trials of EMG data for each subject were analyzed to determine average peak amplitudes for all muscles during each concentric and eccentric phase of each exercise.

Statistical Analyses

The EMG data were collected during both concentric and eccentric phases of both the prone hamstring curl as well as the razor curl, quantified by integration, and expressed as mean electrical activity for each phase of the exercise. Data from each muscle were normalized by being expressed as a percent contribution of the MVIC to the total electrical activity of all 4 muscles tested. Statistical analyses were performed by using SAS 9.1. A paired t-test was used to compare each muscle's peak activation for the razor curl and the traditional prone curl at the p ≤ 0.05 levels.


The paired t-test of EMG data for the medial hamstring, biceps femoris, gluteus maximus, and gluteus medius comparing each muscle's peak activation to the 2 different exercises revealed no significant difference (p ≤ 0.05) (see Table 1). It should be noted that when examining the means of each muscle's percent of MIVC, the razor curl displayed a greater total activation. Figure 6 shows the mean EMG values for each muscle.

Table 1
Table 1:
Differences in peak muscle activation between the razor curl and the traditional prone curl for each muscle.
Figure 6
Figure 6:
A) Medial hamstring electromyography (EMG) data normalized by being expressed as a percent of the contribution of electrical muscle activation of the maximum voluntary isometric contraction (MVIC) for the pull phase of the razor curl (SE = 23.35) and for the concentric phase of the traditional prone curl (SE = 12.59). B) Biceps femoris (lateral hamstring) EMG data normalized by being expressed as a percent of the contribution of electrical muscle activation of the MVIC for the pull phase of the razor curl (SE = 12.13) and for the concentric phase of the traditional prone curl (SE = 14.82). C) Gluteus maximus EMG data normalized by being expressed as a percent of the contribution of electrical muscle activation of the MVIC for the pull phase of the razor curl (SE = 10.28) and for the concentric phase of the traditional prone curl (SE = 11.48). D) Gluteus medius EMG data normalized by being expressed as a percent of the contribution of electrical muscle activation of the MVIC for the pull phase of the razor curl (SE = 7.15) and for the concentric phase of the traditional prone curl (SE = 7.54).


The traditional prone hamstring curl has typically been the exercise of choice when one has been concerned with strengthening the hamstring muscle group. However, when performing the prone hamstring curl, one is not in a functional position nor is one getting the greatest activation of the hamstring when the goal is knee flexion. The position of the prone hamstring curl provides for training the hamstrings concentrically with the hip held in a relatively stable position. The researchers were able to conclude that the prone hamstring curl does, indeed, target the musculature of the hamstrings. However, it has been shown here that the more functional position of the razor curl does, indeed, achieve activation of not only the hamstring muscle group but also the gluteus medius and maximus. More importantly, the razor curl provides one a training method of eccentric hamstring contractions and simultaneous dynamic hip movement. This provides one with a “real-world method” of hamstring functional training in the athletic position as opposed to the prone hamstring curl.

Training in a functional position and training for functionality is critical in sporting endeavors. Ideally one wants to mimic the sporting position as well as the sport demands through the functionality of training. In addition to training functionally, the razor curl addresses the issue of active and passive insufficiency. Biomechanically, active and passive insufficiency deals with a 2-joint muscle such as the hamstrings. The hamstring muscle group is a 2-joint muscle, in that it crosses both the hip and the knee. In order for the hamstring muscle group to have the greatest mechanical advantage at the knee, then the muscle should be place on stretch at the hip. The razor curl places the hip at 90°, thus placing the hamstring on stretch for optimal production of knee flexion. In addition to placing the athlete in a functional position when performing the razor curl, one is also training the hamstrings eccentrically at the knee. Training the hamstrings eccentrically in a position that also provides for gluteal activation provides the athlete the chance to develop neuromuscular conditioning in the functional athletic position. The net effect of this neuromuscular grooving is to provide the athlete the opportunity to place the hip and knee in the optimum position with hamstring activation to protect the ACL.

It can be concluded that both the traditional prone hamstring curl and the razor curl can efficiently train not only the hamstring muscle group but also the gluteals. However, if one wants to promote functionality, one should be training in a more biomechanically functional position as in the razor curl (11). It has been shown that female motor control strategies used during the stop-jump task may, indeed, place the female at a higher risk, but they also suggest that the motor control factors are likely to be the more easily modifiable factor with any of the ACL injury prevention efforts (10,11).

Of all the muscle groups involved, the razor curl exhibited the most mean muscle activation in the medial hamstring. The medial hamstring muscle tendonous insertion is on the anterior tibia. The mechanics of the medial hamstring insertion allow it to play a primary role in resistance to shear anterior tibial forces. The shear anterior tibial forces are the primary culprit to ACL injuries. Therefore, it can be concluded if one attempts to train the hamstrings functionally through the razor curl one may actually be assisting in a form of functional ACL injury prevention, that of which the prone curl does not address.

Practical Applications

It has been shown here, in collegiate female athletes, that, indeed, both the razor curl and the traditional prone hamstring curl both activate the hamstring and gluteal muscle groups. It has also been theorized that the hamstring and gluteal muscle group play an active role in noncontact ACL injuries. Thus, in attempt to reduce noncontact ACL injuries, one should be targeting the hamstrings and gluteals. Both the traditional prone hamstring and the razor curl allowed for hamstring and gluteals activation in the collegiate female athletes who participated in the study. However, if one wants to train the hamstrings functionally, one should focus on the razor curl. The razor curl is designed to increase hamstring contractibility by placing the hip into flexion. Essentially training by performing the razor curl, one accentuates all the other land-based training methods such as jump landing training in efforts to ultimately decrease the susceptibility of ACL injury.

The razor curl can be easily implemented through the use of about any back extension machine. Ideally a coach, athletic trainer, or strength and conditioning coach would have the individual perform first for form and function. Then once the individual has mastered the correct form, a set of 3 sets of 10 would be appropriate. Once the exercise is mastered, the individual should perform the exercise as quickly as possible to mimic the speed of sports performance. After one is able to perform the razor curl at a functional speed with appropriate form, they can then add weight to progress again focusing on form then speed. The uniqueness of the razor curl is the ability to train at functional speeds. Yes, there are other exercises that do address hamstring and gluteals activations as the traditional prone curl does; however, the ultimate question to be answered is which activity is the most functional?


The authors would like to acknowledge and thank Andrey J. Stone for her assistance in the statistical analysis of this paper.


1. Basmajian, JV and Deluca, CJ. Apparatus, detection, and recording techniques. In: Muscle Alive, Their Functions Revealed by Electromyography. Butler, JP, ed. Baltimore: Williams & Wilkins, 1985. pp. 19-64.
2. Hatcher, L. Paired-samples t test. In: Step-by-Step Basic Statistics Using SAS. Cary, NC: SAS Institute Inc., 2003. pp. 451-488.
3. Hewett, TE, Myer, GD, and Ford, KR. Prevention of anterior cruciate ligament injuries. Curr Womens Health Rep 1: 218-224, 2001.
4. Hintermeister, RA, Lange, GA, Schultheis, JM, Bey, MJ, and Hawkins, RJ. Electromyographic activity and applied load during shoulder rehabilitation exercises using elastic resistance. Am J Sports Med 26: 210-220, 1998.
5. Holm, I, Fosdahl, MA, Friis, A, Risberg, AM, Myklebust, G, and Steen, H. Effect of neuromuscular training on proprioception, balance, muscle strength, and lower limb function in female team handball players. Clin J Sports Med 14: 88-94, 2004.
6. Hurd, WJ, Chmielewski, TL, and Snyder-Mackler, L. Perturbation-enhanced neuromuscular training alters muscle activity in female athletes. Knee Surg Sports Traumatol Arthrosc 14: 60-69, 2006.
7. Ireland, ML. The female ACL: why it is more prone to injury? Orthop Clin N Am 33: 637-651, 2002.
8. Irmischer, BS, Harris, C, Pfeiffer, RP, Debeliso, MA, Adams, KJ, and Shea, KG. Effects of a knee ligament injury prevention exercise program on impact forces in women. J Strength Cond Res 18:703-707, 2004.
9. Kendall, FP, McCreary, EK, Provance, PG, Rodgers, MM, and Romani, WA. Muscles: Testing and Function (4th ed.). Baltimore: Williams & Wilkins, 1993.
10. Liu-Ambrose, T, Taunton, JE, Macintyre, D, Mcconkey, P, and Khan, KM. The effects of proprioceptive or strength training on the neuromuscular function of the ACL reconstructed knee: a randomized clinical trial. Scand J Med Sci Sports 13: 115-123, 2003.
11. Oliver, GD and Dougherty, CP. The RAZOR curl: a functional approach to hamstring training. J Strength Cond Res, 23: 401-405, 2009.
12. Weinhold, PS, Stewart, J-DN, Hsin-Yi, L, Cheng-Feng, L, Garret, WE, and Bing, Y. The influence of gender-specific loading pattern of the stop-jump task on anterior cruciate ligament strain. Injury 38: 973-978, 2007.

hamstring strengthening; functional training; ACL injury prevention; functional training

© 2009 National Strength and Conditioning Association