Don,

I looked over the predictions of 18 of the M6.5+ events
claimed as successes by geoF. 18 is all they would
give out from a 5-month period of their 18-month test
yesterday, 6-10/2001.

One glaring problem is that the majority of the quakes
were bigger than the predicted size range, some by two
magnitude units, and as I remember all were at least
1.5 units above the lower bound in magnitude.

Is an accurate warning of an M8+ quake a prediction of
an M4-6 quake? I don't think so. Given the Gutenberg-
Richter distribution, which says there are many more
small quakes than big quakes, none of the 18
predictions I saw would give people confident
knowledge of an impending dangerous quake.

We'd love to be able to predict dangerous earthquakes, and may find a way some day, but haven't yet.

John