Skill Ratings, Part Two

Power, Accuracy, Short Game and Putting skills can be measured. While it may not be an exact science, formulas can be developed to track these and rank players. Can we trust them though?

The Numbers GameThe PGA Tour gave it a first pass, and now I’m going to give it my best shot. Last week we looked at the PGA Tour’s “Skill Ratings” for the members of the Presidents Cup team, and suffice to say I had some issues with how the numbers shook out. This week I’ll put in my two cents on how to rate thes four skills: Accuracy, Power, Short Game, and Putting.

Nothing is going to be perfect. All I’m hoping is that these might evolve over time to become reliable or at the least, believable. Oh, and also you should be able to understand exactly what they mean. Nothing vague. This should be a step up from last week.

Power
Last week, the PGATour.com had driving distance (DD) as the only variable in power. This should probably consist of the majority of the percentage of how to rank players in this category. Not only is it one of the few statistics that provide insight into power, but the best. I won’t even try to dispute it.

What I will add is another and slightly obscure stat: Going for the Green (GFTG). What this will give power is a bit of flavor. GFTG will help rank players who are not only long, but put that power to good use. This is a great way to give a bit of a bump to those guys.

So here is the formula for Power:
Power = (0.75 * DD) + (0.25 * GFTG)

Accuracy
Last week I thought that this was the area that needed the most work. Somehow, PGATour.com gave Tiger Woods a 10 for accuracy. Yes, Mr. Woods is in the top five for greens in regulation (GIR). Yes, GIR has been established (at least in my mind) as the most important statistic in relation to scoring and money. But I’m sorry – Tiger is not a 10. He wasn’t last week and he isn’t this week.

To find the most accurate players I wanted to weigh GIR heavier than any other statistic because:

  1. It IS the most important and
  2. You can’t hit a lot of GIR without being spot on with your irons

This statistic is all about hitting the ball where you want.

After GIR, the next best thing to use is driving accuracy (DA). This was the missing link I was referring to last week that would have taken Tiger’s ranking down a notch (or four). While DA is not as important as GIR, accuracy is not simply a measure of hitting greens. There are, normally, 14 chances to hit fairways in a given round and keeping the ball in play off the tee should be a component of accuracy.

Last, I thought there should be something that helped GIR. I did not want to reward players that hit a lot of greens, but were leaving themselves 50-60 feet from the pin on a regular basis. Proximity is an obscure stat that lets us know just how close the players are hitting it to the pin, right down to the inch. Adding this in will give those guys who seem to dial it in on a regular basis a bit of a boost.

In the end, I came up with this formula for Accuracy:
Accuracy = (0.50 * GIR) + (0.25 * DA) + (0.25 * Proximity)

Short Game
I wish I had access to statistics that told us how close (proximity) players were hitting shots within 75 or 50 yards of the green. This would be one of the numbers that would go into calculating short game rankings. Without that, the next best statistic is scrambling (Scr) which tells us how well players are getting it up and down. I mentioned last week that this is good because it not only combines chipping and bunker play, but putting as well… to get it up and down, the putt must go in.

The other stats going into the equation are sand save percentage (SS), scrambling from rough (SFR) and proximity from sand (SProx). Sand save percentage will count for more of a percentage than the others. I thought that adding scrambling from rough and proximity from sand was a good way of, similar to accuracy, rewarding those a little better with the wedge in tougher spots.

The formula for Short Game:
Short Game = (0.60 * Scr) + (0.25 * SS) + (0.075 * (SFR + SProx))

Putting
Last week, it was very vague how this was calculated. It didn’t even seem that putting average (PA), the stat used to determine the best putter on tour, wasn’t used. I won’t make that mistake. Some people like using putts per round (PPR) to deterimine how well they are putting, but I don’t believe it is nearly important. If you miss a bunch of greens and have a great day chipping the ball close, you can have an abnormally low number of putts in a round. Putting average takes chipping out of the equation.

One other obscure stat that I saw and thought should be included was 3-putt avoidance (3PA). Pretty self explanatory… avoiding a 3-putt is good. Below is the calculation for Putting:
Putting = (0.60 * PA) + (0.30 * PPR) + (0.10 * 3PA)

Thoughts
I think that this is a good start. It may not be perfect, but it’s good. I may tweak these a bit based on some comments and feedback for next week’s column. That will be the last installment of this topic. I’m going to apply the formulas to all these statistics and calculate the rankings in each category for all the players on tour. I have no idea what to expect and it may show us how good or bad these formulas are.

Over time, I’m hoping for access to some more comprehensive statistics that will perhaps better represent the four categories. You have to play with the cards you are dealt though. The question is: Did I play them right? Next week we may find out.

3 thoughts on “Skill Ratings, Part Two”

  1. This was very thoughtful, actually it was a great article but I wanted to come up with something different.
    I think some of these may need a little work but I really enjoyed the article.

    The ones who hit 50 and 60 feet from the hole are still going to get a better score with the + (.25*proximity) make it a minus to take away from their score. OR maybe it is too early.

  2. Dave,

    Thanks for the comments, I really appreciate it.

    You are correct about Proximity needing to penalize the golfer…and it does looks like I am ‘adding’ in the wrong way. But the way the calculation works is that the lowest score will be the best in that skill rating. ‘Proximity’ in the formula is the ranking a person has in that statistic. Here is an example:

    Currently Adam Scott is ranked 1st so his total for it will be (1 * .25) = .25. On the other end, Ben Crane is ranked 149th so his total will be (149 * .25) = 37.25. So when we add the other stats in the same manner a cumulative total and a ranking and skill rating.

    You can find a good explanation of exactly how each stat is calculated at the bottom its page on PGATour.com. Proximity is found at http://www.golfweb.com/stats/leaders/r/2005/331.

    I hope this clears it up. Again, thanks for reading. If you have any other comments/questions, go ahead and post them.

Leave a Reply

Your email address will not be published. Required fields are marked *