In sports such as basketball, players frequently have to make reactive or feinting movements against opponents. In particular, defensive players often have to continue facing their opponents to cutoff dribble drives or to prevent them from passing or shooting the ball. Studies have shown that male basketball players spend 31% of game time performing sliding or shuffling movements, 20% of which is spent performing high-intensity sliding movements (14). Studies have also shown that most acute lower extremity injuries occur during sudden deceleration motions or changes in movement direction (2,7,17). Therefore, it is important to clarify methods for safely and effectively performing lateral deceleration-acceleration motions such as lateral cutting maneuvers from sliding.
To perform effective lateral cutting maneuvers from sliding, players should be able to decelerate and accelerate their body center of mass quickly (i.e., with short foot contact time) and produce a fast lateral center of mass velocity (12). Thus, when assessing a player’s ability to perform lateral deceleration-acceleration motions such as cutting maneuvers from sliding, both the fastness (body center of mass lateral velocity) and the quickness (foot contact time for decelerating and accelerating the body center of mass) should be assessed.
Frontal plane hip neuromuscular functions or biomechanical parameters such as hip abductor strength or hip abduction are generally considered to be important factors determining the capacity to perform efficient lateral motions such as sliding (15). However, whether hip frontal plane functions determine the capacity to perform efficient lateral movements of the body center of mass remains unclear. For example, investigation by Kea et al. (10) of the relationships between the frontal hip abductor or adductor strengths and the lateral jumping distances during one-leg lateral hops revealed a poor correlation. Researchers thus concluded that hip abductor or adductor strengths are not necessarily determining factors to increase lateral jumping distance. To the best of our knowledge, insufficient scientific information exists on how frontal plane hip neuromuscular functions or biomechanical parameters influence the capacity to perform lateral cutting maneuvers.
However, recent studies have found much clearer relationships between the sagittal plane lower extremity biomechanical parameters and the jump height. Ford et al. (8) investigated how hip and knee extensor moments and joint angles during drop jumps could be associated with the jump heights of adolescent women. They reported that more than 80% of the variance in jump heights could be predicted using these variables. Similarly, Lees et al. (13) examined the contributions of the ankle, knee, and hip joints to the submaximal and maximal vertical jump performance. Their study showed that the hip joint work increased significantly with the vertical jump effort, whereas the ankle and knee joint works remained unchanged. Collectively, these studies indicate the importance of the hip sagittal plane joint, especially the hip joint, in propelling the body center of mass in the vertical direction during jump tasks. Accelerating the body center of mass laterally can also be performed by simply changing the angle of pushing (or kicking) the ground using sagittal plane joint motions. Thus, it can be considered that not only the frontal plane but also the sagittal plane hip neuromuscular functions and biomechanical parameters might play important roles during lateral motions such as sliding or lateral cutting maneuvers.
Currently, insufficient scientific information exists on the contributions and importance of frontal and sagittal plane hip biomechanical parameters on the fastness and quickness of lateral cutting maneuvers. Clarifying such information should provide important background knowledge for developing strength and conditioning training programs and for coaching players in defensive play.
Thus, the current study aims to investigate the relationship between the frontal and the sagittal plane hip biomechanical parameters and the fastness and quickness of lateral cutting maneuvers from sliding motion. We examined the following hypotheses: (a) faster hip abduction and extension velocities at the foot contact of lateral cutting maneuvers is related to faster and quicker lateral cutting maneuvers; (b) larger ground reaction force and lesser angle of ground reaction force vector are related to faster and quicker lateral cutting maneuvers; and (c) the lesser the height of the body center of mass during a lateral cutting maneuver, the lesser is the angle of the ground reaction force and the faster and quicker is the lateral cutting maneuver.
Experimental Approach to the Problem
Subjects were instructed to align the lateral edges of their left foot to a line that was the length of their height away from the center of the force plate and take a base stance for a defensive move (Figure 1). They were also instructed to perform lateral sliding and cutting maneuvers as they would actually perform during defensive play in basketball. Thus, they generally positioned their upper body upright or with only slight flexion relative to the vertical so that they could continue to face in front throughout the task. From this position, they were instructed to take 2 steps toward the left, then make lateral cutting maneuvers, and step back to the starting point as fast and quickly as possible. Before actual data collection, the subjects practiced the lateral cutting maneuver until they could perform the series of sliding-lateral cutting maneuver-sliding motions at their usual levels. Each participant performed the maneuver 3 times. Data were discarded if the participants made the maneuver when even part of the foot was outside the force plate or if they felt that they could not perform the task at usual levels.
Twenty-eight healthy, noninjured female college basketball players without history of orthopedic surgery to lower extremities or meniscal injuries participated in the study (age: 20.7±1.0 years, height: 166.0±8.5 cm, and body mass: 58.7±7.5 kg). All subjects read and signed the University Institutional Review Board-approved consent form before participating in the data collection procedures.
All biomechanical data collections were performed in the afternoon in July and August 2009. All subjects wore their own basketball shoes during the biomechanical data collections to be able to perform lateral cutting maneuvers with best performance. All data were obtained using a 3-dimensional (3D) electromagnetic motion tracking system (Motion Monitor System; Innovative Sport Training, Inc., Chicago, IL, USA). The motion sensors were securely attached on the dorsum side of the shoe and directly on the skin above the flat portion of the tibial shaft, lateral aspect of the distal thigh just proximal to the lateral femoral epicondyle and on the iliotibial band, and on the sacrum using double-sided tape, and they were further secured using athletic underwrap and white tape. After the lower extremities and pelvis of the subjects were digitized, the subjects performed lateral cutting maneuvers from sliding. To consider the different lower extremity lengths among participants, the starting point of the lateral cutting maneuver was set at a point that was the length of each subject's height away from the center of a nonconducting force plate (Type 4060; Bertech Corporation, Columbus, OH, USA). All subjects performed lateral cutting maneuvers on the force plate using their left legs after 2 steps of sliding motions. Basketball players must move laterally in both directions during games, and they need to perform cutting motions in all directions and essentially have no dominant leg to perform the lateral cutting maneuver. Therefore, in this study, we only analyzed data for the left leg.
Data Collection Procedures
The electromagnetic tracking system collected position and orientation data for each segment and ground reaction force data during the lateral cutting maneuver with a sampling frequency of 140 and 1,000 Hz, respectively. The position and orientation data were synchronized to the ground reaction force data. Each segment had an orthogonal 3D local coordinate system in which the +y-axis was aligned superiorly and parallel to the longitudinal axis of each segment, the +x-axis was aligned anteriorly, and the +z-axis was aligned laterally for the right and medially for the left leg. Motion Monitor Software was used to calculate 3D joint angles with Euler’s equations in the rotational sequence z y′ x″ (9). For lower extremity joints, proximal segments were referenced for their adjacent distal segment to describe relative joint angles. All kinematic and kinetic data were processed using a low-pass fourth-order zero-lag Butterworth filter at 12 and 40 Hz, respectively.
Data Reduction and Analyses
The foot contact time during the lateral cutting maneuver was defined as the period from the moment that the vertical ground reaction force (GRFv) exceeded 10 N to the moment it became zero after foot contact. We used the sacrum body center of mass to represent the body center of mass (16). Horizontal displacement data of the sacrum body center of mass in the medial and lateral directions relative to the global coordinate system were differentiated to obtain the instantaneous linear velocity of the sacrum center of mass, and this velocity was considered representative of the instantaneous linear velocity of the body center of mass in the lateral direction. To evaluate the fastness and quickness of the lateral cutting maneuver, a lateral cutting index (LCIndex) was calculated using the following equation:
where VCOM and TFC represent the instantaneous linear velocity of the sacrum center of mass at take off and the foot contact time of the lateral cutting maneuver, respectively.
To assess the first hypothesis, the maximum hip extension and abduction angular velocities immediately before (HExtVMax and HAbdVMax, respectively) and at foot contact (HExtV@FC and HAbdV@FC, respectively) of the lateral cutting maneuver were determined. To examine the second hypothesis, the X (horizontal) component of the maximum ground reaction force (xGRFMax) in the global coordinate system that is directed toward the starting point of the lateral cutting maneuver and the Y (vertical) component of the ground reaction force at xGRFMax (vGRF@xGRFMax) were determined. The angle of the ground reaction force at foot contact at xGRFMax (GRFAngle) was calculated using the following equation:
To examine the third hypothesis, the height of the sacrum center of mass from the force plate in the vertical direction was determined at 3 points: (a) at initial foot contact in the lateral cutting maneuver (SCOMHt@FC), (b) at the transition moment when the direction of lateral velocity of the sacrum changed (SCOMHt@TM), and (c) at take off in the lateral cutting maneuver (SCOMHt@TO). These sacrum center of mass heights were expressed as a ratio of each participant’s body height to normalize the difference in height across participants. To increase the reproducibility of the biomechanical data, the values of each variable from 3 successful lateral cutting maneuvers were averaged to determine representative values for each participant (4).
We initially calculated all kinetics and kinematic variables based on the aforementioned local and laboratory coordination systems using the right-hand rule. However, to facilitate interpretation of the results relative to the hypotheses, we assigned the hip extension and abduction, xGRFMax, toward the starting point, and sacrum lateral velocities at take off in the lateral cutting maneuver as positive. For each hypothesis, simple regression analyses were conducted to predict LCIndex with kinetic and kinematic variables. Then, stepwise multiple regression analyses were conducted to collectively examine these relationships. The alpha level was set at 0.05.
Descriptive statistics for each kinetic and kinematic variable are presented in Table 1. HExtVMax and HAbdVMax occurred 22.8 ± 21.7 milliseconds (range, 2–103.7 milliseconds) and 82 ± 45.1 milliseconds (range, 5.33–213 milliseconds) before foot contact, respectively. The relationship between LCIndex and HExtVMax (Figure 2A) showed a significant positive correlation with LCIndex (R2 = 0.166, B = 0.0054, p = 0.03). The relationship between LCIndex and HExtV@FC (Figure 2B) showed a similar but nonsignificant correlation with LCIndex (R2 = 0.128, B = 0.0039, p = 0.06). The relationship between LCIndex and HAbdVMax (Figure 3A) showed no significant correlation with LCIndex (R2 < 0.001, B = 0.001, p = 0.994). Interestingly, the relationship between LCIndex and HAbdV@FC (Figure 3B) showed a significant negative correlation with LCIndex (R2 = 0.160, B = −0.005, p = 0.04). These results indicated that subjects with greater LCIndex demonstrated greater maximal hip extension velocities immediately before foot contact and lesser hip abduction velocities at foot contact.
The relationship between LCIndex and xGRFMax (Figure 4A) showed significant and strong correlation between these variables, indicating that the greater xGRFMax was, the greater LCIndex became (R2 = 0.539, B = 1.863, p < 0.001). The relationship between LCIndex and vGRF@xGRFMax (Figure 4B) showed the same but relatively weaker correlation with LCIndex (R2 = 0.296, B = 1.285, p = 0.003). In contrast, the relationship between LCIndex and GRFAngle (Figure 4C) showed a significant negative correlation between these variables, indicating that the lesser GRFAngle was, the greater LCIndex became (R2 = 0.343, B = −0.193, p = 0.001). These results indicated that participants who showed greater xGRFMax and vGRF@xGRFMax as well as smaller GRFAngle demonstrated greater LCIndex.
The relationship between LCIndex and SCOMHt@FC (Figure 5A) showed a significant negative correlation between these variables, indicating that the lesser SCOMHt@FC was, the greater LCIndex became (R2 = 0.353, B = −35.873, p = 0.001). Similarly, both SCOMHt@TM (R2 = 0.228, B = −25.027, p = 0.01) and SCOMHt@TO (R2 = 0.224, B = −30.091, p = 0.01) showed a significant negative correlation with LCIndex (Figures 5B, C, respectively), although these correlations were slightly weaker than that between SCOMHt@FC and LCIndex. Collectively, these results indicated that subjects who lowered their body center of mass during lateral cutting maneuvers demonstrated faster and quicker lateral cutting from sliding.
To collectively analyze the relationships among sagittal and frontal plane technical factors and the performance of lateral cutting maneuvers, we conducted a multiple stepwise regression analysis to predict LCIndex using SCOMHt@FC, HExtVMax, and HAbdV@FC. These predictor variables were used because they showed more significant correlations with LCIndex than did other similar kinematic variables. In addition, the horizontal GRF is the only force that pushes the body center of mass laterally during the lateral cutting maneuver, and it must be directly related to LCIndex. Therefore, a stepwise regression analysis to predict xGRFMax using the above kinematic variables was also conducted.
When predicting LCIndex using SCOMHt@FC, HExtVMax, and HAbdV@FC, SCOMHt@FC was entered into the equation first; it was found to account for 35.3% of the variance in LCIndex (p < 0.01). Although only a tendency was seen (p = 0.088), when HExtVMax was entered into the equation in the next step, it was found to account for an additional 7.3% of the variance in LCIndex, and the following equation was obtained:
When predicting xGRFMax, HExtVMax was entered into the equation first, and it was found to account for 23.7% of the variance in xGRFMax (p = 0.009). Then, only SCOMHt@FC was entered into the equation, and it was found to account for an additional 11.2% of the variance in xGRFMax (p = 0.049); as a result, the following equation was obtained:
The current study investigated how frontal and sagittal plane biomechanics are associated with the fastness and quickness of lateral cutting maneuvers from sliding in female basketball players. The results suggested that players with greater xGRFMax, GRFv@, xGRFMax, and HExtVMax and lesser GRFAngle, HAbdV@FC, and SCOM@FC had greater LCIndex.
Lowering the body center of mass by flexing lower extremity joints is typically considered important in basketball to move quickly (12), especially during lateral movements. Modulating the vertical body center of mass may influence the frontal and sagittal plane hip biomechanics and other important kinetic parameters such as the amount and direction of ground reaction force during lateral cutting maneuvers.
In basketball, players have to always recognize changes in movements or places of other players or the ball and react to them as quickly as possible. Specifically, during defensive play, the ability to quickly change the direction of momentum of the body center of mass is essential. LCIndex is calculated by dividing the lateral velocity of the body center of mass at take off by the foot contact time. This means that LCIndex reflects both how fast subjects reduced the momentum of the body center of mass in the lateral direction to zero and produced the momentum of the body center of mass in the opposite direction during the cutting maneuver. Thus, it is considered that LCIndex reflects a subject’s capacity for power output in the lateral direction during the lateral cutting maneuver as the performance index evaluating power production in the vertical direction during a drop jump (3,18). Therefore, LCIndex is considered to appropriately evaluate the fastness and quickness of performing the lateral cutting maneuver from sliding in the definitions of the current study.
In this study, subjects were required to make a lateral cutting maneuver with short foot contact time and return to the starting position as fast as they could. To do so, subjects needed to receive enough ground reaction force to decelerate and accelerate their body center of mass by powerfully kicking the floor during the lateral cutting maneuver from lateral sliding. Thus, it was considered important for them to move the hip joint quickly before foot contact so that they could powerfully kick the floor. We considered whether the sagittal or frontal movement is more important when kicking the floor to receive a ground reaction force high enough to result in faster and quicker lateral cutting maneuvers. Therefore, we first investigated the relationship between LCIndex and the hip extension and abduction velocities in terms of the maximum values immediately before foot contact and at foot contact.
In support of our first hypothesis, our results indicated that greater HExtVMax was significantly related to greater LCIndex; HExtV@FC showed a similar tendency. Surprisingly, greater HAbdV@FC was significantly related to lesser LCIndex; HAbdVMax showed no significant relationship with LCIndex. Furthermore, these hip abduction variables were not significant predictors for LCIndex in the multiple regression analyses. These results indicate the importance of hip extension motion to powerfully extend the hip before foot contact in lateral cutting maneuvers. HAabdVMax might not be related to LCIndex because the directions of frontal plane hip motions in around half the subjects were adductions instead of abductions at foot contact. These results may indicate that hip abduction motions are not highly associated with kicking the ground for fast and quick lateral cutting maneuvers. Thus, our results partially support the study by Kea et al. (10) that showed poor relationship between the hip abductor strength and the jumping distances during one-leg lateral hops.
The amount of horizontal ground reaction force received during foot contact in the lateral cutting maneuver was considered to be directly related to the velocity of the body center of mass after foot contact. The angle of the ground reaction force in the frontal plane should also be important to obtain an adequate amount of horizontal ground reaction force during the lateral cutting maneuvers. Our results showed that greater xGRFMax and lesser GRFAngle were significantly associated with greater LCIndex. Furthermore, as hypothesized, our results showed that a lower-body center of mass during lateral cutting was associated with higher LCIndex. We consider these results to be closely associated and support the general suggestion that basketball players should keep flexing their lower extremities to lower their body center of mass with a wide stance, particularly during defensive moves.
Because players implement lateral cutting maneuvers using one side of their leg, the ground reaction force vector during cutting must run from the center of pressure of the foot toward the body center of mass (Figure 6A). If the ground reaction force vector does not run through the body center of mass at take off, the body will rotate around its center of mass (Figure 6B) because it produces a moment (or torque) around it. The body center of mass is located at approximately 58.81% of the person’s height from the foot on average (5), that is, it is approximately located at one’s navel. When instructing players on how to make a lateral cutting maneuver, the angle between the line connecting the foot used to make the cutting and the navel and the horizontal line can be a useful indicator for evaluating whether players receive the ground reaction force at a proper angle.
It should be noted, however, that greater vGRF@xGRFMax is also significantly associated with greater LCIndex, although the extent of association is relatively weaker than that between xGRFMax and LCIndex. This positive relationship can be explained by the principle of frictional force. The frictional force is the product of the frictional coefficient and the normal force (compressive force between 2 surfaces). Thus, excessively small GRFAngle may result in insufficient amount of normal force (vGRF) to receive sufficient frictional force (xGRF) to make fast and quick lateral cutting maneuvers. Therefore, if players try to excessively minimize GRFAngle, their feet would slip during cutting maneuvers.
The aforementioned physical implications about the amount and direction of ground reaction force may lead to questions about the appropriate angle to increase LCIndex. The GRFAngle (corresponding to θ in Figure 6A) of the participant who had the highest and lowest LCIndex was 49 and 62 degree, respectively, that is, the range is 48–62 degree. These values may suggest that coaches should instruct players to maintain a cutting angle of ∼50 degree between the floor and the line between their foot and navel. It should be noted, however, that because frictional coefficients differ somewhat depending on the condition of the floor or material of the sole of shoes, coaches and players should instinctively determine the best kicking angle to realize fast and quick lateral cutting maneuvers.
Collectively, these results indicate that lowering the body center of mass and powerfully kicking the ground with faster hip extension motion before foot contact may be important for implementing fast and quick lateral cutting maneuvers from sliding and for receiving higher ground reaction force. Thus, the current results do not support the notion that improving the hip abductor strength is important for performing efficient lateral motions (15).
This study has some limitations. First, the subjects were all female college basketball players without previous knee orthopedic surgery or meniscal injuries. Thus, the obtained results may be applicable only to similar populations. Second, we only discussed the importance of hip extension and abduction motions from the viewpoint of performing the lateral cutting maneuver. Many studies have shown that strengthening hip abductor muscles is beneficial for reducing knee pain and improving lower extremity functions (1,6,11,19), and all tasks were conducted in anticipated situations, which is not always applicable during actual games or practice situations. Therefore, our results should only be applied to restricted scenarios such as a training program for improving the defensive moves of basketball players. Third, all subjects wore their own basketball shoes, presumably with different friction coefficients between the shoe and the force plates. However, we instructed participants to perform the lateral cutting maneuver as fast as they could, and their feet never slipped when making lateral cuttings. Thus, it was thought that the maximum frictional forces (xGRFMax) they received were equal amount to or below the maximum static frictional forces, and therefore, wearing their own basketball shoes was thought to minimally affect variables (i.e., xGRFMax, GRFAngle, LCIndex) during the lateral cutting maneuver. Finally, the current study only used correlational analyses that do not entirely show cause-and-effect relationships. However, we have discussed physical and theoretical explanations of why these relationships might be established; thus, we believe that the relationships we showed can, at least in part, explain cause-and-effect relationships. In the future, an intervention study should be conducted to confirm whether the relationships found in the current study are cause-and-effect relationships or simply coincidental associations.
Our results suggest that both hip extensor muscles and lowering the body center of mass are important for performing quick and fast lateral cutting maneuvers from sliding. These results have some practical implications. First, players should train to have sufficient lower extremity muscle extensor power output capabilities, strength, and endurance to kick the floor and keep the body center of mass sufficiently low throughout the game to move quickly in lateral directions, say, during defense. Second, any potential factors that limit the lower extremity range of motion to lower the body center of mass (e.g., limited dorsiflexion range of motion) should be modified to perform faster and quicker lateral cutting maneuvers from sliding. Third, because quickly moving the hip joint while lowering the body center of mass is considered important, it may be beneficial to practice stepping exercises such as ladder training by lowering the body center of mass rather than by raising it. Finally, coaches may use the angle between the line between the player’s kicking foot for the lateral cutting maneuver and the navel and the horizontal line as an indicator to determine whether players are making lateral cutting at proper angles.
We are very grateful to Miss Aki Kinjo and Miss Ayumi Higashi, the assistant coach and trainer, respectively, of the Osaka University of Health and Sport Sciences basketball team to which our subjects belonged, for their valuable advice regarding the experimental settings and the organization of subjects’ schedules during the data collection sessions. In addition, we greatly appreciate the critical advice received from Dr. Travis Ficklin concerning the physical concept discussed in the present study.
1. Baldon Rde M, Lobato DF, Carvalho LP, Wun PY, Santiago PR, Serrao FV. Effect of functional stabilization training on lower limb biomechanics in women. Med Sci Sports Exerc 44: 135–145, 2012.
2. Boden BP, Dean GS, Feagin JA Jr, Garrett WE Jr. Mechanisms of anterior cruciate ligament injury. Orthopedics 23: 573–578, 2000.
3. Bruhn S, Kullmann N, Gollhofer A. The effects of a sensorimotor training and a strength training on postural stabilisation, maximum isometric contraction and jump performance. Int J Sports Med 25: 56–60, 2004.
4. Chappell JD, Yu B, Kirkendall DT, Garrett WE. A comparison of knee kinetics between male and female recreational athletes in stop-jump tasks. Am J Sports Med 30: 261–267, 2002.
5. Clauser C, McConville J, Young JW. Weight, Volume and Center of Mass of Segments of the Human Body, in: AMRL Technical Report. OH: Wright Patterson Air Force Base, 1969.
6. Ferber R, Kendall KD, Farr L. Changes in knee biomechanics after a hip-abductor strengthening protocol for runners with patellofemoral pain syndrome. J Athl Train 46: 142–149, 2011.
7. Ferretti A, Papandrea P, Conteduca F, Mariani PP. Knee ligament injuries in volleyball players. Am J Sports Med 20: 203–207, 1992.
8. Ford KR, Myer GD, Brent JL, Hewett TE. Hip and knee extensor moments predict vertical jump height in adolescent girls. J Strength Cond Res 23: 1327–1331, 2009.
9. Kadaba MP, Ramakrishnan HK, Wootten ME, Gainey J, Gorton G, Cochran GV. Repeatability of kinematic, kinetic, and electromyographic data in normal adult gait. J Orthop Res 7: 849–860, 1989.
10. Kea J, Kramer J, Forwell L, Birmingham T. Hip abduction-adduction strength and one-leg hop tests: Test-retest reliability and relationship to function in elite ice hockey players. J Orthop Sports Phys Ther 31: 446–455, 2001.
11. Khayambashi K, Mohammadkhani Z, Ghaznavi K, Lyle MA, Powers CM. The effects of isolated hip abductor and external rotator muscle strengthening on pain, health status, and hip strength in females with patellofemoral pain. J Orthop Sports Phys Ther 42: 22–29, 2011.
12. Krause JV, Meyer D, Meyer J. Basketball
Skills & Drills. Champaign, IL: Human Kinetics, 2008.
13. Lees A, Vanrenterghem J, De Clercq D. The maximal and submaximal vertical jump: Implications for strength and conditioning. J Strength Cond Res 18: 787–791, 2004.
14. McInnes SE, Carlson JS, Jones CJ, McKenna MJ. The physiological load imposed on basketball
players during competition. J Sports Sci 13: 387–397, 1995.
15. Roozen M. Developing hip joint adduction and abduction strength. NSCA Perform Train J 4: 18–19, 2005.
16. Saini M, Kerrigan DC, Thirunarayan MA, Duff-Raffaele M. The vertical displacement of the center of mass during walking: A comparison of four measurement methods. J Biomech Eng 120: 133–139, 1998.
17. Shimokochi Y, Shultz SJ. Mechanisms of noncontact anterior cruciate ligament injury. J Athl Train 43: 396–408, 2008.
18. Taube W, Leukel C, Lauber B, Gollhofer A. The drop height determines neuromuscular adaptations and changes in jump performance in stretch-shortening cycle training. Scand J Med Sci Sports 22: 671–683, 2012.
19. Willy RW, Davis IS. The effect of a hip-strengthening program on mechanics during running and during a single-leg squat. J Orthop Sports Phys Ther 41: 625–632, 2011.