I think as long as the 3 essential parameters are presented, the predictor should be able free to do what he/she wants.

A 3.9-4.1 prediction is more exact than a 3.5-5.5 prediction. If a 4.0 quake occurred, I would give more credence to the 3.9-4.1 prediction than I would with the 3.5-5.5 prediction. The narrower the prediction, if hit, is more NOTEWORTHY it is.

Michael