If you’re not a “math guy”, this may be of little interest. Unless that is, you’re interested in reading a simple “how to piece” on how to manipulate your handicap. Before I get started, let me point out that there are those – on and off the Handicap Committee – that don’t believe the “numbers” can reveal anything. What follows is an example of what the numbers can tell us. You be the judge. And yes, this is very real data for a very real golfer at Gainey. It may have been moved to slightly to mask the identity of this golfer, but it is real.
This particular member plays a fair amount of his golf on courses other than his home course. The chart below shows his posted differentials for a substantial (read: statistically significant) period of time. The blue diamonds show his differentials when playing on Gainey Ranch G.C. The red squares are his posted differentials when playing at away courses.
A couple of things immediately jump out at you. Home differentials are clearly randomly distributed around roughly thirteen. The actual number isn’t the important thing; it’s the distribution that is significant. You can see they range from a low of around seven to a high of roughly twenty. If you analyze the distribution, it is what you would expect for someone in this golfer’s handicap range, i.e., standard deviation of around three strokes. Ninety-five percent of the differentials should (given the laws of statistics) fall between approximately seven and nineteen. What an amazing coincidence – they do!
Now look at the away scores. Their average is closer to sixteen, nearly three strokes higher than those recorded at Gainey. Does this fact imply handicap manipulation? Maybe – maybe not. It can certainly be argued that Gainey Ranch’s handicap doesn’t “travel well”. I’m confident there’s an element of truth to that. You can also make a case that a golfer isn’t as familiar with away courses and a lack of course knowledge results in higher scores. Perhaps this is true, but how many times can that excuse be used? After all, once you’ve played a particular course four or five times, you should have a pretty good level of familiarity and that excuse tends to evaporate into the morning mist.
For the sake of this discussion, let’s temporarily accept that a three stroke difference in the scoring average is acceptable. If that’s the case, shouldn’t the entire grouping of data points move up approximately three strokes? For this golfer, it does, but . . .
The “randomness” disappears! It’s as if the entire bottom half of the distribution is missing! Could it be that the scores weren’t posted “properly”? Hell yes. That’s one possible explanation. The standard deviation of away postings is approximately one stroke! This seems to be statistically “improbable” (for those of you not paying attention, this is called a gross understatement).
An average differential of sixteen with a standard deviation of one stroke means that ninety-five percent of this golfer’s away differentials fall within the range of fourteen and eighteen. This is well within the “you must be joking” range of statistical probabilities.
There are possible explanations. For example, it could be that by playing a more difficult course with which you have no familiarity whatsoever magically makes you a much steadier and more consistent golfer than you are at home. To me, that’s like saying the more you drink, the better you drive.
There are a couple of other explanations. I’ll sit back and see if any of you come up with them and post them as comments here. In the meantime, I continue to do battle with those who say numbers don’t prove anything and I’ll watch to see how this golfer performs in the upcoming Member/Member tournament. This golfer isn’t alone. He has company. I’d love to say, “If you don’t mind, I don’t mind”, but I do. It’s my job.