For those of you following my recent lists/posts, I've been fiddling with game(r) clusters a lot lately. The 5-cluster grouping that I posted about in my other recent list ( http://www.boardgamegeek.com/geeklist/18879) seems to work pretty well.

While some games strongly and primarily match one cluster or another, others are more spread between clusters. Further, if you take a particular gamers Top 10, noone is going to exclusively match one cluster. The results of these matches are a little complex to interpret, so I made a way to visually represent it, and I think it's very neat.

It's a "spider plot"/"cobweb plot"/"radar plot", of the quality of match of each of the 5 clusters where the clusters have been ordered in a way that seems to make sense. The size of the "web" corresponds to the match quality.

In this way, you can generate a "fingerprint" for a game, a gamer, a game group (by entering the top/most played games of that group), a designer, or any number of other game related things. I've posted a few. The tool to make the graphs is at: http://mkgray.com:8000/cgi/gamercluster2