You all might want to check the Syzygy Evaluation board.

Mike there has come up with a method of predicting quakes that is so statistically significant it's off the chi-square tables.

Unfortunately, it's global quakes, no locations, so it's essentially useless for warnings but mathematically, it's fascinating.

He uses a window based on the overlapping time between a 7 day new moon window and a 7 day perigee window. The overlap portion is the window for predictions and he gets WAY too many hits to be chance.

Roger