Additionally, it should be determined in advance what period of time after a quake which fulfills time and location parameters has its magnitude assigned.

As we all know, it is often days, weeks or even months before a "final" figure is determined. The early adjustments are sometimes rather large. This additional step (re-examining magnitudes later) might be applied only to prediction series which appear to be successful.

Which brings up a point I'm not sure has been discussed. How many fulfilled predictions should be required before a prediction series is deemed successful?

Mike W.