Given its size, GCA is a pretty homogenous group. Most of us are looking for the same things in golf courses, and then we squabble over the minor details. This thread is about the minor details.
Using the data from Ian Linford’s unofficial GCA’s top courses project, I have tried to determine the factors that separate some of us from others. Using a statistical technique called “factor analysis,” I have tried to order different raters along preferential dimensions that predict their preference for different types of courses. I’m having a difficult time figuring out exactly what these dimensions represent, and I would greatly appreciated your help.
If it helps, these dimensions are comparable to the ideological dimensions of Congressmen who might have one ideology on moral issues and another ideology on economic issues. Even if we don’t know a Congressman’s ideology, we could potentially tease it out by observing her roll-call votes. In the same way, I am observing each GCAers course ratings and trying to tease out where they fall on these preferential dimensions.
There are two dimensions that seem to predict the preferences of individual raters above and beyond the average quality of a course. The first, strongest dimension seems to represent individual preferences for brand-name, classic-era courses. Raters on one end of this dimension show a strong preference for Pebble Beach, Merion, Pine Valley, Cypress Point, Ballybunion, TOC, Royal Dornoch, Royal County Down, and Shinnecock. Those on the other end of this dimension show a stronger (than expected) preference for Kingsbarns, Olympic, Old Head, Bel-Air, Butler National, Chicago, Yeamans Hall, Lost Dunes, LACC, and Blackwolf Run. It appears that some of us strongly prefer the big-name courses while others tend to favor hidden gems and more modern designs. Do you have any further suggestions about what this dimension could be picking up.
The second dimension confuses me even more. On one end of this dimension, raters like Beverly, Princeville, Bethpage, Oakland Hills, Tralee, Whistling Straits, Casa de Campo, Blackwolf Run, Skokie, and Pinehurst No. 2. On the opposite end, people prefer Walton Heath, Sunningdale, Pasatiempo, Winged Foot, Royal Cinque Ports, Quaker Ridge, Pine Needles, Old Town, and Woodhall Spa. This dimension could pick up different preferences for public vs. private, links vs. parkland, modern vs. classic; but none of those seem to fit perfectly. I need some help from the DG in figuring this out. What is it that separates these two sets of courses?
Even though this is a little bit confusing, I think these factors really do predict an individual’s preference for different types of courses. I have played 5 courses on the list since I submitted ratings for Ian’s project, and these factors correctly predicted in 4 out of 5 cases whether I would like a course more or less than the average GCAer. Moreover, these factors can help to predict my preference for courses that I have not played. For example, this analysis is telling me that I should seek out Ballybunion, Beverly, and Royal Liverpool over Kingsbarns, Loch Lomond, or Butler National. For another GCAer, the recommendation might be exactly the opposite.