Chris;

I'm fumbling around trying to get a significance in the absence of standard tests.

I have 226 of Don's CA predictions. Since he has small lower limits I had to do CNSS searches to cover his ranges. But the file I use for probability estimates also was for larger quakes so I had to redo that too.

Finally I could check his predictions against the quakes looking for matchups in all 3 parameters, finding 33 such hits. Then I did a Jones calculation for probability.

But his predictions all have different odds and there's no test for significance in such cases. I tried using the average probability as an estimate but I know it's wrong because the range is very large.

Roger