I wrote about course form in general here.
However, I realized I didn’t really look into how to adjust the value of a player properly for this week. There are two factors that I think are real, if not very important: Experience on course(# of rounds), and whether the course suits a players skills*.
*As I have stated, these are very small. You can come very close to evaluating a player’s chances to win by just 2-year rankings. But, I’m going to try to get a little closer based on course history.
I think it would be pretty dumb to overlook the benefits of playing in a tournament multiple times. Knowledge of the course and greens and even things like familiarity traveling or staying in the same hotel are definitely things that could affect a players value. This is the first part of defining any course form.
I used the graph that compared each players number of rounds played at Firestone to their results relative to two-year average. Then I adjusted for the overall average of field. Doing this, I came up with a number to add to the players score for number of rounds played.
It ranged from -.08 for Phil Mickelson who has played 64 rounds on Firestone’s South Course to .02 for every player who has never played a round at Firestone*.
It’s important to remember that compared to the field this turns into around -.05 to Phil, and +.05 for zero rounds. The numbers above just came out of the equation I used, but when compared to the field turn into range from -.05 to .05. That’s about a range of .3 strokes per round.
This part gets a lot harder. The first thing I need to do is take out any effect that experience might have. So, I took course form and adjusted it based on the number of rounds played. So, Tiger, based on his performance at Firestone can be expected to be about -.1 standard deviations better here, but since he’s played 44 rounds that accounts for about .04 of it. That means Tiger gets about a .06 bonus for his actual skills relating to the course*.
*At this point I’m assuming that Tiger any relationship is related to skill and not just variance. There is about a 5% chance that a player of Tiger’s caliber would play that far above average in one random tournament over 44 rounds. So, this could just as easily be randomness.
The second thing we need to do is standardize all the course performance numbers. For example, Hunter Mahan in 4 rounds has an average of -.77 which is significantly above his average. That gives him a huge bonus in my equation, but which is more impressive: Tiger who has played to about a -1.13 average(given strength of field, that’s probably .3 above average for him) in 44 rounds or Mahan, who had one good week? Obviously, Tiger.
In four rounds a players average can vary much more than in forty. So I took a players relative score to average, than standardized it based on the standard deviation of averages based on the rounds they had played. Tiger was around -2.7 standard deviations better than average performance. Mahan was only around -1.5. By doing this we can get a better feeling for how each player actually achieved at Firestone.
Finally, I took all that, through it into a few equations and here are the Odds I came up with for the week:
|Miguel Angel Jimenez||0.6380%||15573|
COURSE EXP AND FORM ADJUSTED
|Miguel Angel Jimenez||0.5187%||19178|