I don't know why I didn't see this before, but it dawned on me that Alan is using all of California for his pseudo predictions of M3.3+ events for each week.

Based on a FAQ that is periodically posted to the earthquake newsgroups, an M3.5+ event with a 7 day window has a 81.6993% probability of occurring by chance. An M3.0+ with a 7 day window has a 98.7287% probability of occurring by chance.

The URL for the web site containing the above FAQ and numbers is:

http://www.deja.com/=hg/[ST_rn=bg]/getdoc.xp?AN=457562651&CONTEXT=922812258.669384710&hitnum=116

This FAQ contains probabilities of earthquakes occurring by chance for 3 separate areas as follows:

- CENTRAL COAST RANGES, CALIFORNIA
- NORTHERN AND CENTRAL CALIFORNIA
- CALIFORNIA

They give the probability for events beginning with M3.0 in .5 increments up to M7.5 for differing periods of time.

The probabilities I listed above were for California M3.5+ events using 7 day windows.

Note also that the listed probabilities on the FAQ are based on a declustered catalog meaning that things like aftershocks were deleted from the catalog prior to doing any calculations.

Based on the above, it is pretty easy to see that forecasting M3.3+ events with a 7 day window for all of California is like predicting the sun will rise everyday. It doesn't say anything.

If I remember right, Alan was doing this because of Berklands predictions. The problem is that Berkland isn't posting predictions for all of California. It might seem like he is, but the fact remains that Jim posts two predictions that
are for different areas in California that still aren't for all of California.

So after going thru all of this, I'm not seeing what the purpose of Alans' postings are since nobody is making a prediction for M3.3+ events covering all of California with a 7 day window.

Did I miss something somewhere?

Dennis