Abstract:

The researcher examined the two major professional golf associations, the Professional Golfer’s Association (PGA) and the Ladies Professional Golfer’s Association (LPGA), to determine physical characteristics relevant for success. The researcher found that those players born outside of the U.S. consistently earn more money and have lower average scores in the most recent professional season. These results are consistent across both tours. The researcher attempted to uncover individual statistical categories that influence this finding. He found that players born outside of the U.S. have significantly superior putting averages, while there appears to be no significant difference in other categories, such as driving and hitting greens in regulation. The superior performance of players born outside of the U.S. remains after controlling for these statistical areas.

Introduction:

Many studies have examined the skills necessary to succeed at the game of golf. Among the first were Davidson and Templin (1986), who found hitting greens in regulation and putting to be the most important determinants of success. These results have been consistent in numerous studies, including Jones (1990), Shmanske (1992), Belkin, Gansneder, Pickens, Rotella, and Striegel (1994), Nero (2001), Dorsel and Rotunda (2001), and Engelhardt (2002).

This paper took a different approach. The researcher examined the impact of professional golfers’ physical characteristics on performance, measured by the money earned and the scoring average for the 2006 season. The researcher examined both the Professional Golfers Association (PGA) and the Ladies Professional Golfers Association (LPGA) tours. There has been considerable evidence that talent and skill level in professional golf have been increasing rapidly over the past decade (e.g. Chatterjee, Wiseman, and Perez, 2001). This is likely attributable to the rapid improvement in equipment, but just as much, if not more, is attributable to increased participant abilities. The researcher examined whether natural characteristics such as size influence professional success in both tours.

Further, the researcher examined the effect of the individual’s origin, specifically segmenting those born in the U.S. and those that were not. Following the two most recent Ryder Cups, which the European teams won handily, there has been much discussion on the “advantages” they must have. Those advantages appear to be more than statistical, as the American team fielded the top three players in the world during the most recent Cup and were still unable to threaten the lesser-known European team. It has been suggested this is potentially due to the team camaraderie the international team enjoys and the American team lacks. While this is a question the researcher cannot answer, he attempted to determine whether this “international effect” is evident in settings other than these group competitions.

This work makes numerous contributions to the literature related to this subject. First, to his knowledge, the researcher is among the first to directly examine the individual physical characteristics of professional golfers. The researcher took an opposite approach than most studies do. Characteristics such as putting and driving are explanatory variables, which the researcher used in an attempt to explain the variation found in relation to the physical characteristics.

Second, the researcher contributed to the literature (e.g. Wiseman, Chatterjee, Wiseman, and Chatterjee, 2004) that examines gender differences in professional golf. While the researcher primarily examined both tours separately, he found the results to be consistent for both tours, suggesting an overall effect, rather than a gender-specific anomaly. In addition, the researcher examined the 2006 professional season, which has just recently concluded. Given the increased quality of equipment, as well as the improved performance of the participants, it is important to examine the most recent data.

The researcher found relations to be consistent with those documented in past literature. In addition, the most interesting finding dealt with the nationality variable. The researcher found those individuals born outside the U.S. scored lower and earned more money than those born within, a result consistent across both professional tours. Further, by examining the primary performance-predicting variables, the researcher found players born outside of the U.S. have lower putting averages, while the other performance-predicting variables are statistically equal. Therefore, it appears that putting explains some of the variation between U.S. born and non-U.S. born players. However, the researcher also examined end-performance controlling for the predictive statistics (including putting average) and found the significant relation remains. Therefore, there appears to be undefined influence. The researcher briefly discussed possible reasons for this, for example, prior experience on international tours.

Literature Review:

Davidson and Templin (1986) were among the first to examine the characteristics that are important in golf success. Examining 1983 PGA data, they concluded that relative to driving, skills of finesse, such as putting and hitting greens in regulation (GIR) were more important statistical areas in relation to performance, as measured by money earned and scoring average. The results suggest that golfers who possess proficiency in many shot-making areas have a higher probability of success than those players with proficiency in a few.

Numerous studies have examined the same topic and have concluded the same, that putting and GIR are the most important determinants of success. Among these are Jones (1990), Shmanske (1992), Belkin, Gansneder, Pickens, Rotella, and Striegel (1994), Wiseman, Chatterjee, (1994), Engelhardt (1995,1997), Moy and Liaw (1998), and more recently Nero (2001), Dorsel and Rotunda (2001), and Engelhardt (2002). This finding is not exclusive to professional golf, as Callen and Thomas (2004) found that amateur golfers must possess a wide array of shot-making skills to be successful, particularly putting and hitting GIR.

Several studies have also examined the incremental significance of certain statistical areas for male golfers in comparison to female golfers. For example, Wiseman, Chatterjee, Wiseman, Chatterjee (1994) examined the PGA, LPGA, and SPGA tours and found that males drive the ball farther and hit more GIR. Consistent with the above results, they also found the most important characteristics for LPGA golfers are putting and greens in regulation. Moy and Liaw (1998) found the same, adding that PGA participants perform better in sand saves relative to the LPGA tour. Shmankse (2000) noted that the PGA tour yielded a superior putting average relative to the LPGA.

Further, Nero (2001) estimated golfers earnings based upon driving distance, driving accuracy, putting average, and sand saves. He concluded that professional golfers would benefit by improved putting more than increased driving distance. Callen and Thomas (2006) extended their previous study by examining male and female NCAA amateur golfers. They reached two primary conclusions: (1) males and females possess different levels of shot-making skills, and (2) these disparate skills influence tournament performance differently across genders. These disparate skills are consistent with those found in professional golf.

Moy and Liaw (1998) asserted that men’s larger physical size and superior strength explained the advantage enjoyed by professional male golfers over their female counterparts, as they can drive the ball farther. However, others have argued that successfully driving the ball requires more than just strength. For example, Hume, Keogh, and Reid (2005) analyzed both driving and putting and found that strength is certainly important in both areas, but flexibility and timing are also critical for success. The related hypothesis is that gender-related differences are therefore related to one or more of those physiological areas. Myers, Gebhardt, Crump, and Fleishman (1993) found statistical support for this; male golfers score higher in strength and stamina, while females have superior flexibility.

Data and Methods:

All data in this study are available online. The researcher obtained PGA and LPGA statistics from the official websites, www.pgatour.com and www.lpgatour.com, respectively. For each tour, the researcher obtained end-performance measures and performance-predicting measures. The researcher’s primary measures of end performance were total money earned during the year (Money) and scoring average (Scoring). Money is defined as the sum total earnings due to end tournament placement throughout the entire season.1 Scoring is defined as the average score (i.e. number of strokes) obtained through each round (i.e. 18 holes).

Also, the researcher examined four performance-predicting statistics. The first, driving distance (DrivingDist), measured the total length of each participant’s average drive. During each round, two holes were selected to be measured, with special care taken to ensure the holes face in opposite directions to counteract the effects of wind. Drives were measured at the point they come to rest. The researcher also examined driving accuracy (DrivingAcc), which is the percentage of time a player hit the fairway with his/her drive, the first stroke taken on par 4 and par 5 holes. These two variables are included to examine the overall “power game” of each player. Past studies have typically found these variables to be less important than the finesse areas of the game. However, there have been studies (e.g. Engelhardt, 1995) that suggest a reversal in recent seasons, as the game of golf has become a more distance-demanding sport. The researcher attempted to see if this was indeed the case.

The researcher examined each player’s percentage of GIR, defined as having any part of the ball touching the putting surface in two or less strokes than par for each hole. Finally, the researcher examined putting average (PuttAv) for each participant. PuttAv is the average number of putts used only on those greens hit in regulation. Using this measure eliminated biasing the results due to chipping the ball close to the hole and having a relatively short putt. Both of these variables have been found to be significant in relation to performance in almost all studies. Therefore, the researcher attempted to see if this relationship still existed.2

Rather than using the value of each of these performance variables, the researcher chose to use the ranking. In each tour, the participants are ranked based upon each statistical category. In fact, those are the numbers most often quoted when commenting on the various statistics. The researcher chose to use rankings in order to have a consistent relationship between the coefficient signs and each variable. Otherwise, each would have to have its own interpretation. For example, a lower putting average is a positive statistic, while a higher percentage of greens in regulation is a positive. By using the respective rankings, the researcher could consistently say that a higher ranking is a positive signal, regardless of the variable.3

The researcher’ primary contribution was to extend the analysis to control for personal characteristics. Therefore, the researcher also identified several natural physical characteristics. He identified the age (age) of each participant, defined as the number of whole years from the individual’s birth date to the end of the 2006 professional year.4,5 Also, he defined height (height) and weight (weight) to control for the player’s physical structure. Height is measured in inches; weight is measured in pounds. The researcher did not have data on the LPGA player’s weight; therefore, this variable was defined only for the PGA sample. Finally, the researcher identified each individual’s birth place. Using this, the researcher created USA, which is a dummy variable equal to one if an individual was born in any of the 50 states, zero otherwise. As such, the researcher could examine the difference between performance of USA-born players, both in end performance and in performance-predicting variables. All personal characteristics are available online. After excluding those players for which complete data was unavailable, the final sample consisted of 196 PGA professionals and 166 LPGA professionals.

The researcher initially examined summary statistics of both sub-samples. Consistent with Chaterjee, Wiseman, Chaterjee, and Wiseman (1994), the researcher found the unsurprising result that PGA players drive the ball farther than LPGA players. This is consistent with physiological studies, such as Myers, et al. (1993.) However, the researcher found no significant difference between the two tours in relation to greens hit in regulation. He found the putting average on the PGA tour to be significantly lower than on the LPGA tour, consistent with Schmanske (2000). PGA tour players underperformed LPGA players in driving accuracy, also consistent with Myers et al. (1993.) More important to the study, the researcher found PGA players are older, on average. It appears that there is a higher percentage of American born players on the PGA tour relative to the LPGA tour.

Table 1 – Summary Statistics:

The following table represents summary statistics for the sample, segmented by observations from the Professional Golfers Association (PGA) tour and the Ladies Professional Golfers Association (LPGA) tour. For the PGA tour, Age is defined as number of whole years from the individual’s birth date to November 6, 2006, the day after the end of the Tour Championship. For the LPGA tour, Age is defined as the number of whole years from the individual’s birth date to November 15, 2006 the date following the ADT Championships. Height is the individual’s height in inches. Weight is the individual’s weight in pounds. EventNumb is the number of events each individual participated in during the 2006 year. USA is a dummy variable equal to one if the player was born in any of the 50 United States, zero otherwise. Money is the total amount of prize money awarded to each individual during the 2006 tour year in each respective tour. Scoring is the average 18-round score for each individual. DrivingDist is the average number of yards for each drive. During each round, two holes are selected to be measured, with special care taken to ensure the holes face in opposite directions to counteract the effects of wind. Drives are measured at the point they come to rest. DrivingAcc is the percentage of time the player hits the fairway with their drive. GIR is the percentage of the time the player hits the green in regulation (i.e. when the ball is on the green and the number of strokes taken is two or less than par.) PuttAv is the average number of putts used on those greens hit in regulation.

PGA LPGA t-statistic
N 196 166
Personal Characteristics
Age 35.76 30.80 10.82
Height 71.55 66.28 21.76
Weight 180.85
EventNumb 25.78 19.93 11.72
USA .75 .51 4.90
Performance Characteristics
Money 1,188,709.48 252,329.11 10.82
Scoring 71.11 72.89 -16.06
DrivingDist 289.40 250.87 39.47
DrivingAcc 63.41 69.52 -9.66
GIR 65.13 64.58 1.33
PutAv 1.78 1.83 -13.97

Results:

The researcher began by estimating the following regression model via traditional Ordinary Least Squares:

Depi = α + β1EventNumb + β2Height + β3Weight + β4USA + β5Age + εi (1)

where Depi is either Money, Scoring, DrivingDist, DrivingAcc, GIR, or PuttAv. Each variable is the rank of the golfer in the respective tour in each statistical category. Therefore, he could interpret each positive coefficient as a negative effect on end performance, as it indicated a higher value for independent variable will result in a higher ranking value for the performance measure. EventNumb is the number of full-field events the participant entered into during the 2006 season on each tour, and it was used to control for variation that is a result of frequency of play. The results are presented in Tables 2 and 3. Table 2 presents the results for the end performance variables, Money and Scoring. By first examining the PGA results, he found that older players make less money. More interesting, American born players score higher and make less money on average than their counterparts. The coefficient indicates that, on average, non-American born players rank 32 and 34 places higher than American born players in money earned and scoring average, respectively.

Table 2 – Multivariate Analyses: Cumulative Performance:

The following table presents results from the following model:

Depi = α + β1EventNumb + β1Height + β2Weight + β3USA + β5Age + β6LPGAdum+ εi

where the dependent variable is either Money (in columns 1, 3, and 5) or Scoring (in columns 2, 4, and 6). For columns 1 through 6, each dependent variable is the rank of the individual observation in each of the pre-mentioned categories, as defined in Table 1. In columns 7 and 8, the dependent variable is an adjusted rank, segmented by quintiles, where the rank is 1 through 5. LPGAdum is dummy variable equal to one if the individual is on the LPGA tour, zero otherwise. All other variables are as defined in Table 1.

PGA LPGA Total
(1)
Money
(2)
Scoring
(5)
Money
(6)
Scoring
(7)
Money
(8)
Scoring
Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat
Intercept 64.18 .41 51.15 .36 211.52 2.79 182.16 2.13 5.82 2.59 5.58 2.44
EventNumb .06 .07 1.59 1.83 -7.38 -12.85 -5.88 -9.06 -.09 -5.74 -.06 -3.79
Height -1.02 -.43 -1.94 -.91 .25 .22 .26 .20 -.03 -.84 -.03 -.98
Weight .21 .74 .55 2.16
USA 31.79 2.97 33.96 3.51 19.73 3.59 19.06 3.08 .82 5.30 .92 5.89
Age 1.36 2.02 .61 1.00 -.21 -.56 -.28 -.67 .02 1.72 .01 1.16
LPGAdum -.34 -1.44 -.21 -.87
N 196 196 166 166 362 362
Adj. R2 .0640 .1107 .5506 .3820 .1566 .1233

The results for the LPGA were consistent with the results for the PGA tour in that non-American born players outperformed their counterparts in both end performance measures. The average increase in ranking was 20 and 19 places for money earned and scoring, respectively. Neither age nor height had any significant relation to end performance.

Although their primary focus was on the two tours separately, the researcher also combined the two tours in a total sample. The difference in the numbers of golfers would create problematic model estimations if the researcher were to simply use the ranking as the dependent variable. He adjusted the rankings by creating quintiles for each performance measure. Therefore, for the total sample, the dependent variable only had 5 values, 1 through 5, where 1 represented those individuals who ranked in the top quintile in that respective category and 5 represented the bottom quintile. In doing this, the researcher assured the rankings were consistent across the two tours. The researcher included LPGAdum, a dummy variable equal to 1 for those individuals on the LPGA tour, zero otherwise. This variable is designed to control for systematic differences between characteristics on the two tours.

The results for the total sample confirmed those found individually in both tours. Specifically, the positive coefficient on USA indicated that across both tours, American born golfers have higher ranking values in both Money and Scoring, which indicates inferior performance. A negative relationship between age and end performance was found in the PGA rankings, but the significance was only marginal, a product of the insignificant relation in the LPGA tour. However, the highly significant and negative relation between USA and end performance was consistent across the two tours and provided an interesting question. It appeared that, of all the physical characteristics the researcher examined, the most important is nationality. This could be a product of many things, some of which the researcher could not examine. For example, it is well known that many players, particularly on the PGA tour, are successful, established players on tours in their native countries prior to participating in the United States. However, the researcher was unaware of any way to fully capture the increased ability attributable to this prior experience.

Regardless, if non-American born players are outperforming, it could simply be due to superior performance in individual areas, which the researcher called performance-predicting characteristics. It was not the researcher’ intent to examine where these skills are obtained, but rather to determine whether evidence supported the existence of superior skill in each statistical area. Therefore, the researcher examined these variables in an effort to “explain” the results of Table 2. Those results are presented in Table 3.

Table 3 – Multivariate Results:

The following table presents results from the following model:

Depi = α + β1EventNumb + β1Height + β2Weight + β3USA + β5Age + β6LPGAdum+ εi

where the dependent variable is either DrivingDist, DrivingAcc, GIR, or PuttAv. For Panels A and B, each dependent variable is the rank of the individual observation in each of the previously-mentioned categories, as defined in Table 1. For Panel C, the dependent variable is an adjusted rank, segmented by quintiles, where the rank is 1 through 5. All other variables are as defined in Tables 1 and 3.

Panel A: PGA (1) (2) (3) (4)
DrivingDist DrivingAcc GIR PuttAv
Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat
Intercept 354.76 2.99 -174.35 -1.23 90.09 .60 149.38 1.01
EventNumb .82 1.12 .12 .13 .88 .94 .40 .43
Height -4.46 -2.48 3.95 1.84 -1.10 -.48 -2.39 -1.06
Weight -.65 -3.04 .39 1.55 .27 1.00 .50 1.89
USA -5.20 -.64 -5.45 -.56 1.22 .12 19.75 1.95
Age 4.60 8.98 -2.17 -3.55 .47 .72 .11 .17
N 195 195 195 195
Adj. R-Sq. .3671 .0998 -.0124 .0209

 

Panel B: LPGA (1) (2) (3) (4)
DrivingDist DrivingAcc GIR PuttAv
Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat
Intercept 436.30 4.15 -472.63 -4.82 179.53 1.83 231.61 2.45
EventNumb -1.02 -1.28 -2.18 -2.94 -4.54 -6.12 -4.27 -5.98
Height -5.48 -3.39 9.33 6.19 -.14 -.09 -1.12 -.77
USA 7.67 1.01 -6.56 -.93 4.22 .60 26.68 3.90
Age .85 1.69 -.52 -1.11 .04 .09 -.22 -.50
N 164 164 164 164
Adj. R-Sq. .0775 .1956 .1906 .2622

 

Panel C: Total (1) (2) (3) (4)
DrivingDist DrivingAcc GIR PuttAv
Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat
Intercept 13.24 5.87 -8.90 -3.85 4.81 1.99 6.86 2.92
EventNumb -.59 -2.52 .69 2.83 -.05 -2.98 -.06 -3.64
Height -.17 -5.68 .19 6.10 -.02 -.50 -.04 -1.29
USA .05 .34 -.03 -1.87 .25 1.54 .74 4.59
Age .06 5.99 -.03 -2.78 .01 1.11 .00 .08
LPGAdum -.59 -2.52 .69 2.83 -.24 -.95 -.36 -1.46
N 360 360 360 360
Adj. R-Sq. .1452 .1037 .0225 .0768

Panel A examines the PGA tour, while Panel B examines the LPGA tour. Panel C examines the combined total sample. The researcher found, unsurprisingly, that younger, heavier, and taller players on the PGA tour hit longer drives. The researcher also found that younger players had less accuracy in their drives, as did taller players (although the significance is marginal.) In column 3, he sought to determine whether or not any of the personal characteristics help explain the percentage of greens hit in regulation. However, the researcher found the natural characteristics have no significance to GIR. In the last column of Panel A, the researcher found non-USA born players have lower (better) putting averages that their U.S. counterparts. Since this is a variable consistently found to be greatly important in golfing success (e.g. Davidson and Templing, 1986), this could explain, at least partially, the variation of USA in end performance.

Turning to Panel B, the researcher examined the performance-predicting variables for the LPGA tour. He found that taller, younger players hit longer drives, again consistent with expectations. However, shorter players drive more accurately. Most interesting, he also found a positive relation between U.S. born status and putting average rank, indicating non-American born players putt more efficiently. Again, this could explain some of the variation found in Table 2. In Panel 3, he examined the total sample. As expected, given the results in the first two panels, taller, younger players hit longer drives than their counterparts. However, taller players hit fewer fairways than shorter players. Also, as expected, the total sample results confirmed that non-U.S. born players have superior ranked putting averages than U.S. born players.

It appears some of the variation in performance unexplained by individual statistical categories may be due to physical characteristics. In regards to nationality, it may be that the superior performance is due to superior putting abilities, as he found no relation to other performance-predicting characteristics. However, he needed to examine the influence of natural physical characteristics in congruence with traditional predictors of end performance. In order to do this, the researcher estimated the following OLS model:

Depi = α + β1DrivingDist + β2DrivingAcc + β3GIR + β4PuttAv + β4EventNumb + β5LPGAdum + β6Height + β7Weight + β8 USA + (2)β9age + εI

where Depi is either Money or Scoring. The results are presented in Table 4. Panel A presents the results for Money, while Panel B presents the results for Scoring. To be consistent with previous studies, the researcher first examined only the performance-predicting variables. Those are presented in columns 1, 3, and 5 of each panel. The results were wholly consistent with those found in the majority of previous studies in that the two most important statistical categories are putting average and percentage of greens hit in regulation. In fact, the researcher found no significance at all in relation to the two driving measures on the PGA tour. However, driving (particularly driving accuracy) appears to be predictive of superior performance on the LPGA tour.

More important to this study, the researcher wanted to see if all of the variation in end performance can be determined by these performance-predicting variables. In other words, does the significance identified in the previous analyses disappear when combined with these more established measures? The researcher examined this in columns 2, 4, and 6 of each panel. The researcher found the results for the PGA tour to be consistent even when controlling for these variables. Specifically, USA maintained significance while GIR and PuttAv also remained highly significant. Therefore, successful players must be proficient at putting and hitting greens in regulation. However, there still seems to be an unexplained contribution from the individual’s nationality that comes from some undefined factor, perhaps prior experience (and success) on a tour in their native country.

The LPGA results are consistent in that the two most important variables for LPGA golfers are also putting and greens hit in regulation. However, there also seems to be a significant effect of driving accuracy on end performance. In both the LPGA sample and the total sample, USA maintains significance.

Table 4 – Multivariate Results with Both Physical and Performance Characteristics:

The following table presents results from the following model:

Depi = α + β1DrivingDist + β2DrivingAcc + β3GIR + β4PuttAv + β4LPGAdum + β5EventNumb + β6Height + β7Weight + β8USA + β9Age + εi

where the dependent variable is either Money (in Panel A) or Scoring (in Panel B). For each statistical category, the variable is the rank of the individual observation. In columns 5 and 6, the statistical variables are an adjusted rank, segmented by quintiles where the rank is adjusted to take on a value of 1 through 5. All other variables are as defined in Tables 1 and 3.

Panel A: Money
PGA LPGA Total
(1) (2) (3) (4) (5) (6)
Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat
Intercept 9.33 .61 -32.81 -.26 -14.67 -2.36 67.29 1.59 .17 .69 .52 .30
DrivingDist -.02 -.26 -.12 -1.20 .07 1.37 .07 1.95 .03 .61 .02 .34
DrivingAcc -.10 -1.12 -.11 -1.19 .12 2.52 .09 2.30 -.02 -.40 -.02 -.49
GIR .58 7.81 .59 7.94 .51 10.18 .41 10.48 .50 11.38 .48 11.10
PuttAv .48 8.11 .45 7.60 .53 13.17 .38 11.42 .43 11.52 .38 10.07
LPGAdum .01 .05 -.06 -.36
EventNumb -.52 -.70 -3.56 -10.56 -.04 -3.77
Height .59 .32 .17 .25 .01 .21
Weight -.21 -.70
USA 20.96 2.49 8.38 2.91 .41 3.55
Age 1.35 2.15 -.15 -.83 .01 1.25
N 195 195 164 164 361 361
Adj. R2 .4137 .4410 .8028 .8953 .5148 .5495

 

Panel B: Scoring
PGA LPGA Total
(1) (2) (3) (4) (5) (6)
Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat
Intercept 2.11 .17 -53.35 -.53 -18.29 -4.27 39.85 1.12 -.26 -1.20 .10 .07
DrivingDist -.05 -.69 .08 -.81 .06 1.71 .06 1.94 .04 1.08 .04 .91
DrivingAcc -.07 -.94 -.05 -.72 .14 4.39 .14 4.28 .04 .95 .05 1.14
GIR .61 9.81 .58 9.64 .59 17.18 .56 16.66 .55 14.21 .54 14.12
PuttAv .49 9.85 .44 9.09 .45 16.51 .39 13.75 .45 13.77 .41 12.44
LPGAdum -.00 -.00 .06 .38
EventNumb .97 1.58 -1.27 -4.47 -.01 -.73
Height -.33 -.22 -.34 -.61 -.01 -.35
Weight .14 .79
USA 23.91 3.48 .30 3.00 .48 4.70
Age .47 .92 -.19 -1.22 .00 .63
N 195 195 164 164 361 361
Adj. R2 .5258 .5657 .8995 .9134 .6258 .6466

Conclusions:

The researcher examined natural physical characteristics of professional golfers on the PGA and LPGA tours. He controlled for performance-predicting statistical measures, namely driving distance, driving accuracy, percentage of greens hit in regulation, and putting average. The researcher found the percentage of greens hit in regulation and putting average to be the most important characteristics of end performance (i.e. success) in professional golf. This is consistent with numerous prior studies. Driving, particularly driving accuracy, appears also to be important on the LPGA tour, but not on the PGA tour.

More important to this study, the researcher found a strong relationship between nationality and end performance. Specifically, U.S. born players have inferior end performance relative to their counterparts. One explanation for this is perhaps the superior putting averages enjoyed by the non-U.S. golfers.

The researcher’ results have interesting implications for professional golf. It is obvious that golf is an international game more now than ever before, particularly in the United States, where the professional prizes are higher than any other country. Recent domination in Ryder Cup has led many to comment on the unexplained advantage European golfers seem to enjoy during those events. While this work takes a broader approach by examining all international born golfers (and not just European ones ), it provides a good starting point in investigating whether superior performance is contingent on nationality. The researcher’ primary objective was to identify whether such a relationship exists and not necessarily to describe its origin. Therefore, future research could be designed to examine the cause, for example training methods or coaching practices.

Endnotes:

1 During the 2006 season, the PGA tour had a total of 48 tournaments, while the LPGA had only 33.

2 In unreported results, the researcher examined numerous statistics, such as sand saves. However, the researcher found no significance in relation to those variables. In order to remain consistent with the previous literature, the researcher chose to examine only the variables that have been consistently used in similar studies.
The obvious assumption is that the incremental difference between each ranking category carries the same weight. In other words, the difference between rankings 1 and 2 is the same as the difference between 2 and 3.

3 While this is a restriction, there is no reason to believe it would bias the result as the pertinent question in an individual sport is performance relative to other competitors. In order to be absolutely sure, the researcher conducted all statistical analyses using the actual number rather than the ranking. All results were qualitatively identical. Results are available upon request.

4 For the PGA tour, the last event concluded on November 6, 2006 while the last event for the LPGA tour concluded on November 19, 2006. There are events on both tours that are not full-field events, meaning that not all players had the opportunity to participate. While there is no reason to believe this would bias the results, for completeness, the researcher eliminated tournaments in unreported results. The final conclusions are unchanged.

5 The researcher also identified a variable labeled experience, defined as the number of years the player has been a professional golfer. However, the variables age and experience were highly correlated (p = .94), therefore the researcher chose to examine only age. However, in unreported results he repeated all analyses replacing age with experience and find the results qualitatively unchanged.

References:

Belkin, D.S., Gansneder, B., Pickens, M., Rotella, R.J., and Striegel, D. Predictability and stability of Professional Golf Association Tour statistics, Perceptual and Motor Skills, 1994, 78, 1275-1280.

Callen, S.J., and Thomas, J.M. Determinants of success among amateur golfers: An examination of NCAA Division I male golfers, The Sport Journal, 2004.

Callen, S.J., and Thomas, J.M. Gender, skill, and performance in amateur golf, The Sports Journal, 2006.

Chatterjee, S., Wiseman, F., and Perez, R. Studying improved performance in golf, Journal of Applied Statistics, 2002, 29, 1219-1227.

Davidson, J.D. and Templin, T.J. Determinants of success in Professional Golf Association Tour statistics, Research Quarterly for Exercise and Sport, 1986, 57, 60-67.

Dorsel, T.N. and Rotunda, R.J. Low scores, Top 10 finishes, and big money: An analysis of Professional Golf Association Tour statistics and how these relate to overall performance, Perceptual and Motor Skills, 2001, 92, 575-585.

Engelhardt, G.M. It’s not how you drive, it’s how you arrive: the myth, Perceptual and Motor Skills, 1995, 80, 1135-1138.

Engelhardt, G.M. Differences in shot-making skills among high and low money winners on the PGA Tour, Perceptual and Motor Skills, 1997, 84, 1314.

Engelhardt, G.M. Driving distance and driving accuracy equals total driving: Reply to Dorsel and Rotunda, Perceptual and Motor Skills, 2002, 95, 164-180.

Hume, P.A., Keogh, J., and Reid, D. The role of biomechanics in maximizing distance and accuracy of golf shots, Sports Medicine, 35, 429-449.

Jones, R.E. A correlation analysis of the Professional Golf Assocation (PGA) statistical ranking for 1988, In A.J. Cochran (Ed.) Science and Golf: Proceedings of the First World Scientific Conference of Golf. London: E &FN Spon. 165-167.

Myers, D.C., Gebhardt, D.L., Crump, C.E., and Fleishman, E.A. The dimensions of human physical performance: Factor analysis of strength, stamina, flexibility, and body composition measures, Human Performance, 6, 309-344.

Moy, R.L. and Liaw, T. Determinants of professional golf earnings, The American Economist, 1998, 42, 65-70.

Nero, P. Relative salary efficiency of PGA Tour golfers, The American Economist, 2001, 45, 51-56.

Shmankse, S. Human capital formation in professional sports: Evidence from the PGA Tour, Atlantic Economic Journal, 1992, 20, 66-80.

Shmankse, S. Gender, skill, and earnings in professional golf, Journal of Sports Economics, 2000, 1, 385-400.

Wiseman, F., Chatterjee, S., Wiseman, D., and Chatterjee, N. An analysis of 1992 performance statistics for players on the U.S. PGA, Senior PGA, and LPGA Tours, In A.J. Cochran and M.R. Farrally (Eds.), Science and Golf: II. Proceedings of the World Scientific Congress of Golf. London: E & FN Spon., 1994, 199-204.