Statistics are very much a part of modern professional sport. In football we have statistics for a teams possession, territorial domination, corners and shots on target, to name just a few. In horseracing we have a vast array of statistics for horses, trainers and jockeys. In tennis we have statistics for first serve percentages, aces, returns of serve, points won at the net and many more. Sport provides fertile ground for statisticians, and golf is no different.

But how useful are these performance statistics? What do they really tell us? Do they give us a genuine insight into the strengths and weaknesses of a players game or are they just meaningless numbers?

Professional golf is littered with statistics

If you take a look at either the US or European Tours official websites, you will see plenty of statistics. The US Tour website is especially fond of statistics, some helpful and some fairly meaningless. Although there is definitely some value to be gained from analysing the more meaningful statistics, they should be treated with caution. Its always worth keeping in mind that famous phrase lies, damn lies and statistics.

Performance statistics can be useful, especially when trying to understand the strengths and weaknesses of a golfers game. For example, they can provide some insight into whether a golfer drives the ball a long way or is a good putter. The trick is to make sure you use the right statistics, interpret the statistics sensibly and dont draw conclusions that arent there to be made.

So which statistics are useful, and which are dangerously meaningless? Lets start with driving statistics.

Driving Distance

The driving distance statistic measures the average number of yards per measured drive. These drives are measured on just two holes per round. Although care is taken to select two holes that face in opposite directions, to counteract the effects of wind, the fact that only two holes are chosen leaves much to chance. Over the course of any one season, a tour professional will be measured on between 150 and 250 drives. Now thats not a large sample and so the reliability of the statistic is somewhat reduced.

As the sample size for driving statistics is so low, the driving distance statistic should only be used to give an indication of a players length off the tee. It should not be used as some precise measurement of a drivers length off the tee. A difference of less than 10 yards is unlikely to be statistically significant. Larger differences may mean something and the statistic is probably most useful in identifying those at the two extremes, namely the very long hitters and the very short hitters.

Driving Accuracy

The other widely used statistic is driving accuracy. This statistic is defined as the percent of time a player is able to hit the fairway with his tee shot. Unlike the driving distance statistic, the driving accuracy statistic is measured on every par 4 and par 5 hole played. Therefore, the sample size for each player for a whole season usually exceeds 1,000 tee shots and this provides a much more statistically significant result.

However, there is a large inherent problem with the driving accuracy statistic, namely the further a ball travels, the more it deviates from a true, straight line. This means that long hitters are always likely to have relatively poor driving accuracy statistics, simply because they hit the ball further.

It is not uncommon to read statements such as player X is wayward off the tee as he hits just 56% of fairways. Such a statement shows a complete lack of understanding of the basic factors effecting good driving. The player could indeed be deemed wayward if he is a short driver. However, if he is a very long driver, then hitting 56% of fairways is a pretty good performance.

Total Driving

The key to understanding a players driving ability is to read the distance and accuracy statistics together. You can generally expect to see an inverse relationship between the two statistics. However, if a player performs well in both statistics, then you can fairly assume that he is a particularly good driver of the ball. Equally, if a player ranks low in both categories, then he clearly has a problem.

The other consideration when assessing a players performance off the tee is to think about his decision-making capabilities. In 2004, Phil Mickelson showed improved form, much of which he attributed to his greater inclination to show restraint off the tee. In prior years, Phil simply bashed the ball off the tee, without giving enough thought to the benefits of controlled, positional play. It is debatable whether driving statistics can pick up this sort of decision-making capability. In theory, better decisions would translate into better performance statistics. However, it is not clear that this is the case and so a wider understanding of the game and the player is helpful when assessing capabilities off the tee.

Greens in Regulation

The greens-in-regulation statistic, which measures the percent of time a player hits the green in regulation, is similar to the driving accuracy statistic in terms of sample size. Indeed, it is even more reliable, being based on every hole, in every tournament. This means an annual sample size of between 1,000 and 2,000 for most golfers.

The GIR statistic is an invaluable measure of how a players long game is performing. However, the common error with this statistic is to assume that it simply measures a players iron play. This is an over-simplification and fails to consider the effect that driving has on reaching the pin in good order. A player who drives the ball well will give himself a much better chance of reaching the green in regulation. Therefore, if that same player rates poorly on greens in regulation, despite being highly ranked for driving, then he is probably not a good iron player. The implication would be that despite approaching the green from good positions on the fairway, the player does not hit the green often enough.

Used alone, the GIR statistic tells us how good a players overall long game is working, including both driving and iron play. Used in tandem with the driving statistics, it is possible to extrapolate some analysis of a players iron play. If you work from the assumption that a good driver should, other things being equal, achieve good GIR statistics, then any deviation from this in reality will provide some indication of a players iron play.

Putting statistics

There are two commonly used putting statistics, namely putts per round and putting average (or putts per greens in regulation). The putts per round statistic is the least reliable of the two as it makes no allowance for how players reach the green. A player who continually misses the green and chips on to the green from short range will usually start putting from closer to the hole. He should therefore achieve a lower number of putts per round. Yet this is as much a reflection of his poor approach play as it is of his abilities with the flat stick.

The more reliable indicator of a players putting prowess is the putting average as it eliminates the effects of chipping close and one putting. As the putting average is recorded on all greens reached in regulation the sample size is large, usually measured on between 750 to 1,500 greens during each season. The statistic should give a fair indication of a players putting form.

However, one word of caution should be noted when using the putting average. The statistic is still affected by the balls initial proximity to the hole, which can never be constant between players. The better a player is at attacking the pin with his iron shots, the shorter the distance of his putts. A player who is more aggressive with his approach play is likely to cover a shorter distance with his putts per greens in regulation than a player who is cautious with his approach play and faces, on average, longer putts for each green reached in regulation. Often, it is the worlds best players, who are focussed on winning, and can afford to be aggressive, that tend to get closest to the pin. Players who struggle to retain their card often cannot afford to take the same level of risk and so play the safer, less aggressive approach shots. The cautious players will therefore face longer putts and will probably rate less highly on putting average statistics. However, this may be as much to do with their approach play as it does with their putting abilities. Overall, it is difficult to gauge exactly how much approach play affects the putting average statistic, but it cannot be denied that it has some effect. Although the putting average is by far our best guide to a players putting ability, it should nonetheless be treated with a degree of caution.

Sand saves and scrambling

An important element of any golfers game is to be able to get out of trouble. In this regard, two statistics have been developed to measure a players powers of recovery, namely the sand save percentage and the scrambling stat.

The sand save statistic is the percent of time a player was able to get 'up and down' once in a greenside sand bunker. Although well intentioned, the problem with this statistic is that it has a very low sample size. Each season, a tour professional is only likely to end up in a green side bunker between 75 and 200 times. The lack of data available makes this statistic fairly unscientific. Furthermore, the statistic does not just measure a players proficiency from the sand, but also his putting ability. The advice is therefore to ignore this statistic.

The scrambling statistic is a better guide to a players powers of recovery. It measures the percent of time that a player misses the green in regulation, but still makes par or better. Although this statistic does include an element of putting skill, it is based on a better sample size. Over the course of a season, each player will face this situation on an average of 400 to 600 times. So, although not purely a judge of chipping and sand play, it nonetheless provides a better basis for judging a players abilities around the green.

The most useful way to use statistics

If whats written above does not make you wary of relying too heavily on performance statistics, then the sand is probably the best place for your head. Performance statistics, if used intelligently, can be useful in indicating a players strengths and weaknesses. However, statistics should never be treated as gospel and simple judgements based on these figures should never be made.

When analysing a players strengths and weaknesses, performance statistics should never be used in isolation. It is better to use performance statistics in a wider context, as just one tool among an array of analysis techniques. Other forms of analysis can provide valuable insights. These include swing analysis, observation, and monitoring quotes from golfers, coaches and other experts. Used in collaboration, these various elements of analysis become much more insightful. Used alone, they can be misleading. So dont chuck the statistics in the bin, set them in context and use them intelligently.