Golf is a popular sport with approximately 30 million participants in North America and 10–20% of the adult population participating in other countries (16). With this notable popularity, it is desirable to understand what leads to and is associated with high-level performance in golf. Clubhead speed and golf handicap are commonly used as efficient ways to quickly evaluate golf performance (5). High levels of technology have allowed for an understanding that golf swing performance is associated with physiological capacity (15), biomechanics (2), and club characteristics (4). These correlations are functionally relevant, because during the golf swing, the lower extremities push against the ground and the whole body acts as a kinetic chain to develop high clubhead velocity to drive the golf ball over a long distance. As a result, it is important to understand how to most efficiently transfer this energy into the ground during the golf swing.
One way to examine energy transfer during the golf swing is to record the ground reaction forces using force platforms (6,12). It has been suggested that to achieve higher clubhead velocity, an attribute of a higher level of performance in golf, a proper weight shift pattern is required to effectively transfer the force from the ground to the superior segments (7,8). Williams and Cavanagh (17) described the general weight shift pattern during different phases of a golf swing. Previous studies have suggested that the timing and weight transfer patterns during the golf swing were more important than simply the magnitude of ground reaction forces (GRF) in detecting golf skill levels (2,13,14). Richards et al. (14) demonstrated that weight distribution ratio from heel to toe at ball contact was able to predict low (LHCP) and high handicap (HHCP) golfers with an accuracy of 85%. Whereas in additional studies, a large and rapid shift in the vertical GRF from trailing foot to leading foot during downswing were associated with greater ball speed and LHCP scores (2,13).
Although the weight shift patterns during the golf swing have been well described, the difference in weight shift pattern between HHCP and LHCP golfers is still not well understood. Previous studies have focused on the maximum vertical GRF during certain phases and the vertical GRF at certain events during a golf swing (2,13) with less attention on the medial-lateral GRF. In addition, it is unknown whether HHCP and LHCP golfers have different timing profiles in maximum GRF during different phases of a golf swing. All of this information may be helpful during retraining golf mechanics by serving as a biofeedback aid in golf-specific movement retraining. The practical question of the current study was whether HHCP and LHCP golfers demonstrated different magnitude and timing of peak GRFs and whether magnitude and timing of peak GRFs could be a potential modifiable factor in improving golf performance. In addition, we wanted to examine the differences between the leading and trailing legs during a golf swing to better understand lower extremity loading patterns throughout the golf swing. It was hypothesized that LHCP golfers would demonstrate greater and earlier timing of peak GRFs during a golf swing than HHCP golfers. It was also hypothesized that trailing and leading legs would have different magnitude and timing of peak GRFs. The findings of this study may assist professional golf instructors in assessing and training their clients.
Experimental Approach to the Problem
The current study compared the GRF profiles between HHCP and LHCP golfers. The experimental design was a cross-sectional controlled laboratory study. Twenty-eight male golfers were recruited based on the handicap level, which provided a group of LHCP and HHCP golfers. Three-dimensional GRF peaks and the timing of the peaks were recorded bilaterally during a golf swing using 2 force plates. The peak GRF and the timing of peak GRF were compared between HHCP and LHCP golfers. The differences between the leading and trailing legs were also examined. All testing was completed in a laboratory. Comparisons were made between the levels of golfers, and between the trailing and leading legs, to better understand what variables associated with weight shift could be beneficial in optimizing golf performance.
Twenty-eight male right-handed golfers with no history of lower extremity injuries limiting sports participation in the preceding 6 months were recruited for the study. The golfers were recruited based on their handicap score and after recruitment were divided into 2 groups based on their performance level (LHCP and HHCP). The 2 groups consisted of golfers with an LHCP (age, 23.8 ± 5.2 years; height, 1.81 ± 0.04 m; mass, 79.8 ± 12.9 kg; handicap, 0–9; n = 15), and golfers with an HHCP (age, 28.6 ± 7.2 years; height, 1.79 ± 0.06 m; mass, 77.5 ± 11.2 kg; handicap 10, n = 13). The study protocol was approved by the Duke Medical Center Institutional Review Board. Subjects were fully informed of the purpose and risks of participating in the current study and signed informed consent before testing.
After completing the initial demographic questionnaire, the motion analysis collection was completed. All of the subjects were allowed to warm-up by stretching and completing approximately 40 practice ball strikes, which took 5 minutes on average.
Motion capture data were collected using an 8 camera system collecting at 240 Hz (Motion Analysis Corporation, Santa Rosa, CA, USA) to track the motion of the golf club to break down the golf swing into multiple phases. To track this motion, 3 markers were placed along the shaft of the golf club. In addition to the motion capture data, GRF data were collected using 2 force plates, 1 for each foot, sampling at 1,200 Hz (AMTI, Watertown, MA, USA). Subjects wore a standard neutral cushioning running shoe (Air Pegasus; Nike, Inc., Beaverton, OR, USA) during testing. Subjects were asked to stand with 1 foot on each of the force plates in a normal golfing posture and to hit a golf ball (2XS Sports; Bentonville, AR, USA) off an artificial turf tee using a standard 5 iron, provided by the laboratory, into a net that was located 6 ft in front of the tee (Figure 1). Subjects were required to complete 7 acceptable trials in which complete ball contact was made so that the ball was launched into the target area of the netting. Subjects were at normal hydration, sleeping, and nutrition intake levels at the time of testing. However, the testing time of day was not controlled in the current study.
Three-dimensional GRF data and the timing of these values were recorded and analyzed bilaterally. The data for each of the 7 trials were processed using Visual 3D (C-Motion, Germantown, MD, USA). The force plate data were smoothed using a fourth-order Butterworth filter with a cutoff frequency of 100 Hz and normalized to body weight. Once the discrete data points were obtained, the average of 7 trials was used for analysis.
To analyze the trials, the golf swing was divided into different phases based on the position of the club: (a) initiation of the swing at the address position to top of the backswing, (b) top of the backswing to ball contact, (c) ball contact to follow through. The top of the backswing was defined as the point when the clubhead marker reached its maximum height and changed direction. Ball contact was defined as the point when the club was vertical. The end of the follow through was defined as the point when the clubhead marker again reached its maximum height. Analysis of the trials was also divided by the leading and trailing leg. Because of the fact that all subjects were right-handed golfers, all references to the leading leg are synonymous with left leg and all references to the trailing leg refer to the right leg.
The variables of interest in the study were related to the timing and magnitude of vertical and medial-lateral GRF during the 3 identified phases of the golf swing. The specific variables of interest were the magnitude of the peak vertical, medial, and lateral GRF and the relative timing (0–100% of golf swing, start to follow through) of these events. These variables were examined for each of the 3 phases of the golf swing as previously defined.
The main focus of current study was to compare the LHCP and HHCP groups to determine the differences that exist between the leading and trailing legs in these 2 groups. Therefore, the interactions between handicap along with side and handicap main effects were the focus of the results and discussion. Because the different roles of the leading and trailing leg, it was expected that inherent side differences would exist and as a result the main effect for side was not a primary focus of the study. A mix model analysis was used that included handicap status (high vs. low) as the between-subject factor and leg (trailing vs. leading) as the within-subject factor. This analysis was conducted for each study variable for each phase of the golf swing. The type I error rate was set at 0.05 as statistically significant. Statistical analysis was conducted using PASW v. 17 (SPSS, Chicago, IL, USA).
On average, the LHCP golfers were younger but had a similar height and mass as the HHCP golfers. In regard to their golf swing, the LHCP golfers took less relative time from the start of the swing to the top of the backswing (46 ± 4% vs. 50 ± 4%) and also took less relative time from the start of the swing to ball contact (62 ± 4% vs. 69 ± 4%). Because of the inherent differences between the leading and trailing legs during the golf swing, the main effects for leg (leading vs. trailing leg) were significant.
Golfers with an LHCP exhibited a different vertical GRF profile when compared with HHCP golfers (Figure 2). There was a significant interaction for the peak vertical GRF for phase 1 and phase 3. During phase 1, the peak vertical GRF in the LHCP golfers was higher on the trailing foot and lower on the leading foot. During phase 3, this relationship was reversed. There were no interactions between leg and handicap status on the timing variables; however, there were differences because of handicap status alone. Throughout all phases of the golf swing, the LHCP golfers exhibited the peak vertical GRF earlier in the golf swing when compared with the HHCP golfers (Figure 3).
Low handicap and HHCP golfers also exhibited differences in the lateral GRF profile. The medial-lateral GRF was acting in the medial direction for the majority of phases 1 and 2 of the golf swing; therefore, the peak lateral GRF in some cases was the smallest medially directed GRF as the GRF was never oriented laterally during these phases. There was a significant interaction between limb and handicap for the peak lateral GRF during phase 1 and phase 3 (Figure 4). The LHCP golfers exhibited a smaller lateral GRF on their trailing foot with a larger lateral GRF on the leading foot during phase 1 in comparison with the HHCP golfers. During phase 3, this relationship was reversed with the HHCP golfers producing 33% lower GRF on each foot. During phase 2, the LHCP golfers had larger peak lateral GRF for the leading foot and smaller lateral GRF for the trailing foot. There was a significant limb by handicap interaction for the peak lateral GRF timing during phase 3. No other statistically significant interactions existed; however there was a main effect for handicap status. On average, the LHCP golfers exhibited an earlier peak lateral GRF during phase 2 in comparison with the HHCP golfers (Figure 5).
No interactions or main effects were observed for the peak medial GRF for any of the phases of the golf swing. In addition, no interactions were observed for timing of the peak medial GRF during any phase of the golf swing. There were, however, main effects for handicap status for the peak medial GRF timing (Figure 6). The peak medial GRF occurred earlier in phase 1 and phase 2 in the LHCP golfers when compared with the HHCP golfers.
Performance during the golf swing has previously been associated with biomechanical differences. The results of the current study add to this rationale and provide the potential for biofeedback-augmented retraining models that incorporate GRF data. The purpose of this study was to compare the characteristics of the peak vertical and medial-lateral GRF and the timing of these variables between HHCP and LHCP golfers in addition to examining the differences between the leading and trailing legs during a golf swing. It was hypothesized that LHCP golfers would demonstrate greater and earlier timing of peak GRFs during a golf swing than HHCP golfers. It was also hypothesized that trailing and leading legs would have different magnitude and timing of peak GRFs.
The findings of vertical GRFs supported the hypothesis of greater peak GRFs in LHCP golfers. Low handicap golfers demonstrated higher vertical GRFs on the trailing foot during phase 1 and demonstrated a higher vertical GRF on the leading leg during phase 3 of the golf swing. Previous investigators have demonstrated that a greater and faster weight transfer from the trailing foot to the leading foot for the vertical GRF was associated with greater ball velocity and an LHCP (2,9,13). Consistent with previous studies, the results of the current study suggest a greater weight transfer from the trailing foot to the leading foot during the downswing in the LHCP golfers.
The findings of vertical GRF also supported the hypothesis of earlier timing of peak GRFs in LHCP golfers. In terms of the timing of the vertical GRF, LHCP golfers demonstrated peak forces earlier throughout the golf swing when compared with the HHCP golfers. Low handicap golfers took less relative time to the top of the swing and to ball contact. The earlier timing of backswing and downswing suggested an earlier overall swing pattern, which was associated with the earlier timing of peak GRF in LHCP golfers. Barrentine et al. (1) have demonstrated that LHCP golfers have a shorter absolute time for the downswing and total swing than HHCP golfers. Different from the study by Barrentine et al. (1), the swing time in the current study was normalized to 100% for all golfers and represented relative timing. Similar to the findings by Barrentine et al. (1), the current study found that the LHCP golfers had relatively shorter time for backswing and downswing as a percentage of the total swing.
The hypothesis of greater peak GRFs and earlier timing of peak GRFs was also supported by different profiles in the medial-lateral GRF between the LHCP and HHCP golfers. Specifically, differences were observed in the magnitude of the peak medial-lateral GRF. Koenig et al. (9) qualitatively described medial-lateral GRF during a golf swing and concluded that a difference was observed in the medial-lateral GRF profiles between high- and low-skill golfers. However, no quantitative analysis was conducted in the study by Koenig et al. (9). In the current study, the higher peak lateral GRF on the leading foot during phase 1 in LHCP golfers might be associated with a consistent upswing and braking motion of the golf club when it reached the top of the backswing. The larger peak lateral GRF for the leading foot during phase 2 could be related to the strong forward swing motion of the golf club. The higher peak lateral force on the trailing foot during phase 3 might be associated with the braking motion of the golf club after impact. In addition, the LHCP golfers exhibited earlier peak medial-lateral GRF when compared with HHCP golfers. Barrentine et al. (1) reported that the absolute timing for trailing foot posterior shear force and leading foot anterior shear after the top of the backswing was earlier in LHCP golfers when compared with HHCP golfers; however, the absolute timing of the medial-lateral force was not significant. In the current study, LHCP golfers had a more rapid change in the relative timing of medial-lateral force direction, which could contribute to the development of higher clubhead speed typical of golfers with elevated performance.
The hypothesis of different magnitudes and timing of peak GRFs between trailing and leading legs was supported. Significant differences were found in most comparisons between the leading leg and trailing leg. The asymmetrical nature of a golf swing would be expected to result in inherent differences in loading between the 2 legs. During the backswing, vertical force was transferred primarily to the trailing leg. The vertical force was then transferred to the leading leg during downswing and reached its peak around impact. Comparisons between trailing and leading legs during the 3 phases demonstrated this general weight shift pattern, with the trailing leg reaching a peak of 0.75 body weight during backswing and the leading leg reaching a peak of 1.0 body weight around ball impact.
The findings of the current study provide information for golf training. Golf-specific exercise programs emphasize that improving key techniques associated with weight transfer may be able to increase the efficiency of the golf swing and thus improve a golfer's performance (10). The timing difference between LHCP and HHCP golfers suggest the potential importance of focusing on timing of the weight shift during a golf training program. Previous studies have found that general timing training can improve golfers' accuracy, clubhead speed, and putting distance control (3,11). Future studies need to examine the effect of timing and physical training on force generation during specific phase of the golf swing and how these affect performance. Future research should examine the short-term and long-term effect of this feedback on golf performance as it relates to clubhead speed and scoring. The current study had several limitations. Data collections were conducted in the laboratory setting with players wearing a standard running shoe and a practice ball being used as a visual target. However, these conditions were consistent across all participants. These changes to golf environment would limit the generalization of the findings to the real world. No drive performance or accuracy variables were collected. It was assumed that LHCP golfers had better drive performance than HHCP golfers in the laboratory setting. The SDs of dependent variables were reported in the current study; however, the repeatability of GRF profiles (3) between HHCP and LHCP golfers was not investigated.
The current study found different magnitude and timing of peak GRFs between HHCP and LHCP golfers and confirmed that the magnitude and timing of peak GRFs could be a modifiable factor in improving golf performance. Thus, it may be relevant to consider both the magnitude of the forces and the timing of these events during golf-specific training. The rapid increase in the availability of biomechanical data allows for an increased number of tools for biofeedback to improve movement and sport-specific skill. The inclusion of sensors that examine vertical forces is currently available in force platforms and in shoe products for less than US $500, which makes the inclusion of these data in movement retraining protocols less of an obstacle. The results of this study suggest that the biofeedback may be beneficial in improving movement efficiency and success of the golf swing.
1. Barrentine SW, Fleisig GS, Johnson H. Ground reaction forces and torques of professional and amateur golfers. In: Science and Golf II: Proceedings of World Scientific Congress of Golf. Cochran J., Farrally M.R., eds. London, United Kingdom: E & FN Spon; 33–39, 1994.
2. Chu Y, Sell TC, Lephart SM. The relationship between biomechanical variables and driving performance during the golf swing
. J Sports Sci 28: 1251–1259, 2010.
3. Doan BK, Newton RU, Kwon YH, Kraemer WJ. Effects of physical conditioning on intercollegiate golfer performance. J Strength Cond Res 20: 62–72, 2006.
4. Egret CI, Vincent O, Weber J, Dujardin FH, Chollet D. Analysis of 3D kinematics concerning three different clubs in golf swing
. Int J Sports Med 24: 465–470, 2003.
5. Fradkin AJ, Sherman CA, Finch CF. How well does club head speed correlate with golf handicaps? J Sci Med Sport 7: 465–472, 2004.
6. Gatt CJ, Pavol MJ, Parker RD, Grabiner MD. Three-dimensional knee joint kinetics during a golf swing
. Influences of skill level and footwear. Am J Sports Med 26: 285–294, 1998.
7. Hellstrom J. Competitive elite golf: A review of the relationships between playing results, technique and physique. Sports Med 39: 723–741, 2009.
8. Hume PA, Keogh J, Reid D. The role of biomechanics in maximising distance and accuracy of golf shots. Sports Med 35: 429–449, 2005.
9. Koenig G, Tamres M, Mann RW. The biomechanics of the shoe–ground interaction in golf. In: Science and Golf II: Proceedings of World Scientific Congress of Golf. Cochran J., Farrally MR, eds. London, United Kingdom: E & FN Spon; 40–45, 1994.
10. Lephart SM, Smoliga JM, Myers JB, Sell TC, Tsai YS. An eight-week golf-specific exercise program improves physical characteristics, swing mechanics, and golf performance in recreational golfers. J Strength Cond Res 21: 860–869, 2007.
11. Libkuman TM, Otani H, Steger N. Training
in timing improves accuracy in golf. J Gen Psychol 129: 77–96, 2002.
12. Nesbit SM, Serrano M. Work and power analysis of the golf swing
. J Sports Sci Med 4: 520–533, 2005.
13. Okuda I, Gribble P, Armstrong C. Trunk rotation and weight transfer patterns between skilled and low skilled golfers. J Sports Sci Med 9: 127–133, 2010.
14. Richards J, Farrell M, Kent J, Kraft R. Weight transfer patterns during the golf swing
. Res Q Exerc Sport 56: 361–365, 1985.
15. Smith MF. The role of physiology in the development of golf performance. Sports Med 40: 635–655, 2010.
16. Theriault G, Lachance P. Golf injuries. An overview. Sports Med 26: 43–57, 1998.
17. Williams KR, Cavanagh PR. The mechanics of foot action during the golf swing
and implications for shoe design. Med Sci Sports Exerc 15: 247–255, 1983.