Hi all;

A proper prediction really should include the probability of success by chance.

It's not difficult.

Once you decide on a location, mag range and date range all you need to do is run a NEIC search on the parameters.

I usually use 1974 thru 2004 for the right size and location. That tells you how many there have been. Divide by 11323 to get the number of quakes per day and multiply by the number of days in your predicted window to get the expected number in such a window. Multiply by 100 if you want a percent.

This isn't a final answer because of clustering. The answer can come out to more than 1 so it isn't a true probability but it's good enough for a rough estimate.

Roger