There is a thread below from Nov 28 to December 3. It has gotten a bit too far down (I posted on it a couple hours ago).

Island daughter said that you should remove multiple quakes that occur on one day (leave one) from the total quakes. That would affect Roger's statistics and make Amit's record a little better.

Island Daughter took Biostatistics in the last couple of years, and is a Bio graduate.

We could be wrong; she just looked at this for about 2 minutes. But Amit's argument that if there were some number of M6+ quakes and they all occured on the predicted day, and none occured on unpredicted days, Rogers' approach would not give 100% success if there were multiple quakes in 1 day. Roger did not address this point in my opinion.

Chris