Burning Man, 2007
Last year I looked at the GCP data over an 8 year period, during the most intense day, the Saturday of the burning. The result was sufficiently strong that I decided to make a formal prediction this year. Here is part of a description from the previous analysis.
The main event is on the Saturday before Labor Day, September 1st, and the location is west coast USA. The time of the actual burning is 9 pm to 10 pm according to the calendar on the Burning Man website. This corresponds to 0400 to 0500 GMT on September 2nd, which was set for the formal event period. The result conforms to the previous trends, though it is only marginally significant, with Chisquare of 3732.835 on 3600 df, a p-value of 0.060 and equivalent Z of 1.554 (uncorrected). The cumulative curve shows some striking deviations early in that period, and has a moderate positive trend for most of the hour.