I came across this rather amusing model for predicting football results using mostly economic data (click on image for full screen):
Now, we all know Brazil aren't going to win the world cup, but most of us would've predicted they'd fare quite well, and possibly win it (my own failed prediction was with Argentina). What's dubious about the algorithm their using is it predicted Serbia to be finalists! How the hell did they arrive at that conclusion? Well, to give you an indication they do discuss some of the factors included in the model. I'll definitely be coming back to this when I've got a spare moment... They did, however, predict Germany would face, and subsequently knock out, England in the last 16.