I disagree.

It is the responsibility of the predictor to state his ranges in advance. You may want to allow a little slack for someone dumb enough to predict an exact location but that's about all.

The question on a prediction is simply yes or no and if yes, how valuable is it based on seismicity.

Your proposal would allow credit for almost anything which happened anywhere after a prediction. For example, a mag. 8 prediction would take as a hit a quake which took place before the prediction was made (negative date range.) And, since 8's are rare, it would score high to boot!

Roger