Forecasting Tournaments: What We Discover When We Start Scoring Accuracy

The article looked like a load of waffle, but it would be nice to see some basic statistical tests applied, e.g. can we reject the hypothesis that everyone is equally accurate?