Hi all;

I took another look at JOB's windows using my own list of great quakes.
The question was what was responsible for tipping it over into the
significant range. Turns out it wasn't the Kuriles at all but the
perigee shift coupled with the longer timespan being used.

Combined NEIC and ANSS great quakes from 1872 to 2007
197 quakes total       197 quakes total
118 syzygy hits        118 syzygy hits
6 perigee hits         perigee shift excluded
124 total hits         118 total hits
probability 0.542      probability 0.542
significance 0.0084    significance 0.0630

As above with Kuriles removed
192 quakes total       192 quakes total
117 syzygy hits        117 syzygy hits
6 perigee hits         perigee shift excluded
123 total hits         117 total hits
probability 0.542      probability 0.542
significance 0.0038    significance 0.0359

While this represents a form of data tampering in that it
was included at a later date, it is at least reasonable
on physical grounds, gravity effect from the moon being
a little stronger at such times.

Roger