Here is a picture of all the disc golf courses in North America. The data for these maps came from the PDGA Course Directory or Disc Golf Course Review.
For something more animated, here are the courses in the U.S. according to the date the course was installed. The first frame is 1976, the last is 2008. All the courses that do not have a date are thrown into the last frame.
Many people have wondered where the "center" of all the disc golf courses is. Some have tried to calculate where it is. These people usually take some sort of average, or compute the place that minimizes the total distance to other courses. Both methods tend to point to somewhere in the middle of the northern part of Missouri, up near the Iowa border.
I thought those methods gave too much weight to the far-flung courses out west and in Alaska and Hawaii. Not that's there's anything wrong with these courses, but why should the courses farthest from the center have the most influence?
Following that line of thinking, I decided to systematically remove from consideration the courses that are farthest from the center. Then, remove those remaining courses that are now farthest from the center. Keep going until you have nothing left but center. It's like finding the middle of an onion.
Ah, you say, but how can you know which courses are farthest from the center without knowing where the center is? Good question. Here's how.
First, I made the presumption that the center of the disc golf universe is somewhere in the U.S. or Canada. (This is convenient, since I don't have coordinates for many courses off of North America.) So, I removed overseas courses from consideration.
What happens next is best explained with a physical model. Put a map of North America on a cork board. Stick a pushpin in for every 9 holes that a course has. Tie one end of a long string off to one side, and start wrapping the string around the outside edges of the pushpins. After each time around, remove any pushpins that the string is touching. (The pushpins for an 18 hole course would be removed after the second time around.) The last pushpin remaining is the center.
That would be Lions Park III in Plainfield, Illinois.
I calculated the Sphere of Influence graph for the disc golf courses in North America. This is supposed to be a way to find "clusters" of points. I thought it might result in some groupings of courses. Maybe to define regions or something. However, it turns out that a lot of courses - from California to the East Coast are in a single cluster (black). This graph was the most difficult, and least obviously useful thing I've done. Data from the May 19, 2009 version of the PDGA Course Directory. The colors don't mean anything, they are just a way of making the different clusters stand out.
I tried several methods of clustering courses, with the idea of creating regions that were easy to remember, and each contained about the same number of courses. I found that the best way was to use state-fulls of courses, and hand-pick the states to go into each region, using a tool which adds up the number of courses in each region. For example, here is a way to group courses into five regions.
