A franchise-related hypothesis. I think I need Bruise's input on this. It has to do with probability. This is the hypothesis: For each casual dining restaurant X, there is a probability greater than 0.9 that another casual dining restaurant Y exists within a 1 mile radius. They are like deer. When you see one, you know another is around. I think the usefulness of this theory is dubious.

I am still thinking about the Franchise 500 . Not related to the Daytona 500. Unless you assign a franchise to each mile of the race. Which would be sort of cool. It is such a great data set. The question is what to do with it. What I'd like to do ideally is not feasible: that is, create a map of the United States based solely on franchise geographic exclusivities. As part of a franchise agreement, usually the franchisor promises that they will not put another franchise within a certain radius of your franchise. If you plotted all of these, you'd get a crazy looking map. Unfortunately, I think that this data is probably kept under lock and key by each of the 500 franchisors. Potentially you can make some assumptions that can get you started though. For example, that each franchisor grants a 5 mile exclusive radius to the franchisee.

Another idea I had was trying to graph the development of franchises geographically. Taking the Franchise 500 and finding out the year and city that the business first started in. Plot those on map and color code by year.

Any other ideas of how we can turn the Franchise 500 into something interesting?