Replies

IIRC he's always made the same four predictions for the week of the full moon, every week, year after year. One of them is a Magnitude 7+ quake "anywhere in the world"; There are an average of 16 of those a year; thus, in any given week, he has about a 30% chance of being right by dumb luck.

The other four predictions are for various gigantic areas of the west coast with a wide magnitude range; the seismological equivalent of predicting for a week in May that "there will be a tornado in Oklahoma."

Overall, basically he's going to get a "hit" on one or more of his predictions by dumb luck over 50% of the time.

BTW, if he's so great, why didn't he predict the Japan earthquake? Or Northridge?

The fallacy is best summed up by David Deming of the University of Oklahoma:

Berkland has been predicting earthquakes since 1974. He was bound to get one or more predictions right, just by chance. This is known as the Texas Sharpshooter's Fallacy. The Texas Sharpshooter randomly shoots holes in the side of a barn. He then walks up to the barn and paints a bullseye around one hole and proclaims himself a sharpshooter.