Kinematic Analysis of Golf Putting for Expert and Novice Golfers
Giorgos Paradisis & Julian Rees University of Wales Institute, Wales, UK
The aim of this study was to identify some of the kinematic parameters used by
expert golfers (less than or equal to 6 handicap) to optimise putting efficiency
and accuracy, and differentiate the putting techniques of elite and novice golfers.
A 2D video analysis (50 Hz) were used to establish whether any differences existed
for selected kinematic parameters of an 8 ft (2.46m) golf putt between expert (n
= 8) and novice golfers (n = 8). Statistical analysis showed that 9 parameters from
the 26 measured proved to be significantly different at the alpha level P < 0.05
between the expert and novice groups. This study assumed that expert golfers were
more competent putters than novice golfers. Whilst this appears to be a reasonable
claim it may not always be the case. Further improvements to the study could be
to incorporate a scoring system to monitor the outcome of individual trials.
In the modern-day game of golf, putting remains the key to shooting low scores,
and the ability to hole putts can turn a good round into a great round. It has been
reported that putting accounts for approximately 40% of all golf shot played (Gwyn
& Patch, 1993). However, golf practice does not seem to balance the percentage out
between the golf swing and the golf putt. Virtually all golfers, regardless of their
level of expertise suffer inconsistencies in putting performance due to the stroke
being a complex and multi-faceted motor process. This is in part due to the fact
that unlike golf driving, in which the need for maximum club head speed at impact
largely determines the body actions that can be successfully employed, success in
putting can be achieved using a variety of techniques (Cochran & Farrally, 1994).
Despite this revealing statistic and the obvious importance of competent putting,
much of the pedagogical literature is based on the observations and anecdotal evidence
provided by top players and coaches. There is a general lack of published scientific
research and information regarding the kinematics of the putting stroke is scarce.
Thus, the aim of this study was to identify some of the kinematic parameters used
by expert golfers (less than or equal to 6 handicap) to optimise putting efficiency
and accuracy, and differentiate the putting techniques of elite and novice golfers.
The study population consisted of right-handed amateur golfers from the University
of Wales Institute, Cardiff (UWIC) and Celtic Manor Golf Club, Newport (CMGC). This
population was separated into two groups based on their playing ability; expert
(EX) (age 23.3 ï¿½ 3.3 yrs, height 1.80 ï¿½ 0.1 m, handicap 2.3 ï¿½ 1.8, experience 9.3
ï¿½ 2.1 yrs; mean ï¿½ s) and novice (NO) (age 26.5 ï¿½ 3.2 yrs, height 1.7 ï¿½ 0.1 m, handicap
25 ï¿½ 2.6, experience 1.5 ï¿½ 0.9 yrs; mean ï¿½ s). The expert subjects had all represented
UWIC in the British University championship and were practising regularly at the
time of testing. Conversely, the novices from CMGC were recreational players who
played on average once a week.
In order to carry out the study it was necessary to set up an artificial putting
surface to allow for maximum control of potential external variables. A flat Astroturf
surface was marked out with white tape (4m x 1.5m) to yield a straight putt with
no break. At one end of the putting surface was a standard golf hole (4 inches in
diameter) and at the other end was a designated marker that ensured that the actual
distance of each putt was equidistant, regardless of an individual's set-up technique.
The order of putts was randomised for each subject until all subjects had putted
once and then the procedure was repeated. This reduced the effect of muscular fatigue
on the putting stroke and any learning effect that would result from continuous
Two-dimensional video analysis was used to capture three trials from each performer.
A Panasonic F'15HS video camera was mounted upon Manfrotto 117 rigid stationary
tripods 5 m away to capture a full field of performance. The camera was positioned
perpendicular to the plane of performance, operating at 25 f/s with a 1/500 s shutter
speed. Nine superficial markers were attached according to the guidelines suggested
by Plagenhoef (1971) on the vertex of the head and both right and left joints (glenohumeral,
elbow, wrist and knee).
Before the commencement of testing each subject was allowed a considerable putting
warm-up and trial period. Firstly, this was to ensure that familiarisation occurred
for the pace and nap of the putting surface, and secondly, each subject needed to
become accustomed to the same ball and putter being used in the investigation. After
the warm-up 2-D video data was collected for each subject performing a series of
putts from a set distance of 2.46m (8 ft). It was explained that the purpose of
the test was to determine an individual's normal putting technique for successful
putts. The order of putts was randomised for each subject until all subjects had
putted once and then the procedure began again. This reduced the effect of muscular
fatigue on the putting stroke and any learning effect that would result from continuous
Co-ordinate digitising was undertaken on an Acorn Archimedes 420/1 microcomputer
equipped with the Kine System software (Bartlett and Bowen, 1993). Generalised cross-validated
quintic spline that has been derived from a program by Woltring (1986) were then
applied to remove random noise. Reconstruction was based on a user-defined 13-point
model. To aid interpretation of results, key moments were introduced in the analysis
to divide the stroke into five phases (Figure 1), as previously defined for qualitative
analysis by Burden et al. (1998).
Figure 1. Typical stick figure sequences at five instants of the golf putt where
kinematic parameters were measured. Ball address (1), Back swing (2), Through swing
(3), Ball impact (4) and Follow-through (5).
The following parameters were calculated: ball position, stance width and wrist
positioning at ball address (BA), and the putter-head horizontal and vertical displacement,
subject's head movement - horizontal and vertical displacement, timing of the BS,
TS, FT and total putt time, angular displacement of the right and left elbow and
also the angle formed by a line joining the left elbow to left wrist and putter
shaft, maximum horizontal linear velocity and time it occurs, at back swing (BS),
through swing (TS) and follow through (FT).
Digitising Reliability: Reliability and objectivity of the digitising process
was established by repeated digitising of one sequence at the same sampling
frequency with an intervening period of 48 h. The limits of agreement method
(Bland and Altman, 1986) was used to compare these repeated digitised sequences
and produced values for the angular displacement of the left elbow (LE) and the
horizontal displacement of the putter head (PH) based on the equation MD ï¿½ 1.96
SD, where MD = mean of differences between repeated digitised sequences and SD =
standard deviation of these differences, as heteroscedasticity correlation was
close to zero. Given these results (Table 1) it was concluded that the digitised
data were reliable and objective.
Results & Discussion
Statistical analysis showed that 9 parameters from the 26 measured proved to be
significantly different between expert and novice players at the 95% level of confidence
The present study's data for hand positioning confirmed a significant difference
(P<0.05). It has been agreed previously that forward hand positioning locks the
wrists into a firmer position, minimising wrist involvement and ensuring that the
individual putts with the larger muscles of the shoulders (Leadbetter, 1997). It
was therefore possible that the potentially more reliable forward wrist positioning
method of the experts resulted from this variable.
There was a significant difference (P<0.05) for mean linear displacement of the
putter head between BA to final BS position. However, in contrast the BC to final
FT mean linear displacement was not found to differ significantly between the groups.
There has not been any definite recommendations put forward for the displacement
of the putter head for the relative phases of the putting stroke. Interestingly,
the novice displacements tended to agree with the pendulum theory proposed by numerous
experts over the years. Their displacements from BA to BS position and BC to final
FT position were approximately equidistant. In contrast the experts tended to exhibit
the modern theory technique where the putting stroke is executed with the BS being
substantially shorter than the FT.
There was a significant difference (P<0.05) for head displacement change between
the BS to BC phase. It maybe concluded that the more important factor was whether
the head was stationary at the point of ball contact. This measurement was however
outside the present study's scope.
Vertical displacement for the putter head during the BS and FT phases was significantly
different (P<0.05) between the two groups. Experts were characterized by shallow
BS and a more pronounced vertical displacement during the FT. In contrast the novice's
vertical displacement during the BS and FT was far greater than both expert phases.
The greater variability found in the novice group during the back swing and follow-through
phases may have resulted from the breaking of the wrists.
Putting (1998) have referred to the fact that to achieve a crisp ball strike
it is desirable to accelerate the putter through the ball hitting area from a slower
BS. Consequently, as acceleration is a derivative of velocity it would be expected
that the greatest velocity attained would occur at or just after the BC phase. This
occurred for the expert golfers (range, 1.05 -1.3 s) but was somewhat more erratic
for the novice golfers (range, 0.3 -1.2 s), therefore a significant difference (P<0.05)
resulted. However, because all putts analysed were successful they must have been
struck at similar velocities through the BC area. Consequently, as expected no significant
differences were apparent between the actual maximum horizontal velocities attained.
The novice group was however more erratic in the smoothness and velocity pattern
during their stroke.
Recommendations for the relative timing of the phases under investigation do
not directly exist. However certain authors (Leadbetter, 1997) refer to a slow BS
and then an accelerating phase into and through the hitting area. Both groups were
characterised by this general trend, the experts to a greater degree than the novices,
therefore no significant differences existed. The only significant difference (P<0.05)
that did occur was for the FT phase, however, this is not thought to be of any great
significance towards putting technique.
The main limitations of the study were the homogeneous nature of the selected
groups (expert and novice golfers) and therefore any post hoc justifications towards
other groups were problematical and indeed limited. Although the collection of data
in the controlled environment (golf practice bay) had advantages from the viewpoint
of methodological considerations, it nevertheless, did not accurately simulate the
real external or competitive situation. Finally, the analysis of only one putt per
analytical investigation may not always be a true reflection of an individual's
technique, and the validity of using a single performance trial per subject as being
representative of generalised performance outcomes must be questioned (Bates et
This study assumed that expert golfers were more competent putters than novice
golfers. Whilst this appears to be a reasonable claim it may not always be the case.
Further improvements to the study could be to incorporate a scoring system to monitor
the outcome of individual trials.
- Bartlett, R.M. and Bowen, T.M. (1993). Kine System User Guide. Alsager: Crewe and
Alsager Faculty of The Manchester Metropolitan University, 3-27.
- Bates, B.T., Dufek, J.S. and Davis, H.P. (1992). The effect of trial size on statistical
power. Medicine and Science in Sports and Exercise, 24 (9), 1059-1068.
- Bland, J.M., and Altman, D.G. (1986). Statistical methods for assessing agreement
between two methods of clinical measurement. Lancet, 307-310.
- Burden, A.M., Grimshaw, P.N. and Wallace, E.S. (1998). Hip and shoulder rotations
during the golf swing of sub-10 handicap players. Journal of Sports Sciences, 16,
- Cochran, A.J. and Farrally, M.R. (1994). Science and Golf II: Proceedings of the
world scientific congress of golf. London: E & FN SPON.
- Gwyn, R.G., and Patch, C.E. (1993). Comparing two putting styles for putting accuracy,
Perceptual and Motor Skills, 76(2), 387-390.
- Leadbetter, D. (1997). Positive Practice. London: Collins Willow
- Plagenhoef, S. (1971). Patterns of Human Motion: a cinematographic analysis. New
Jersey: Prentice Hall, Inc.
- Putting Dr. (1998). Putting School.
- Woltring, H.J. (1986). A Fortran package for generalised cross-valedictory spline
smoothing and differentiation. Advances in Engineering Software, 8, 104-113.