Amit;

If you are using a 1 day window, predicting the exact date of the quake, the odds are 16/365 or .0438.

But your question is too simple. Even with odds that small it is possible to get more hits than expected. The question then becomes how likely is it that your answer is solely due to chance.

My test is strict because I want to avoid false positives. Thus I want results at the 99% level. The number of hits and misses depend on the number of predictions. If you make 16 predictions you need to get 8 hits to pass.

Roger