My name is Jason, and I'm a student here in Berkeley. I did a statistical project for my Geology and Statistics courses, concurrently. I also discussed some conclusory ideas generated by my study with the head of the Psychology Department. The project was an attempt to assess the statistical accuracy of non-traditional earthquake prediction. Some of the conclusions (in colloquial form) are presented below. I hope this helps the folks who read these boards to keep an appropriate perspective. (The Psychology professor said the people who run these boards are like the Iranian Ayatollahs and will not permit any dissent, and are afraid of the open-air of free discussion. The Professor said my postings will be deleted or defaced, posthaste. We shall see.)

I hope you folks don't mind, but for the past 3 months I've kept a daily chart of predictions on the Syzygy, Pinpoint, and Earthwaves boards. I assume from reading virtually all the postings you site visitors and contributors are all interested in unbiased assessments of your important work in this critical field ... so many lives depend on it.

I correlated all the predictions with actual temblors. I allowed for the vagueness of many of the predictions. Here are the results:

Bob Shannon (whose site is also particularly engaging and informative) was at 2.1% during this period. This seems forgivable since he is focusing on a narrower, less day-to-day active geographical area. He falls into the “wishful-thinking” trap of many predictors who confine themselves to the area in which they live. Since all who predict are on some level rooting for earthquakes to occur, my psychology professor says they will regularly over-predict in hopes of getting more right. Unfortunately this means they will enhance their accuracy arithmetically (at the ideal) but increase their errors geometrically, at least!

Diane (who I really enjoyed on the TV show) had an accuracy reading of 3.4%. Frankly, given her self-promotion, I was shocked at how ineffective she has been, but I did double-check my figures. She seems so convinced of her heaven-sent powers of prediction that she is blinded to the simple truth that she is almost never accurate and seems to rate her own accuracy with ex post facto rationalized miscalculations. I hope she will be candid enough to re-examine her self-assessments of accuracy. Using her line of “thinking”, perhaps God has decided to make her delusional. It troubles me that many of her warnings may be heeded by the naïve. My guess is if God were telling her when the quakes would come He would have a better batting average. Talking baseball metaphors, I would think God’s batting average would be closer to Babe Ruth’s than Doctor Ruth’s. (I do not mean to speak flippantly about God, my father is a Church Elder. But as he told me yesterday when we discussed this, it is hard not to blaspheme when dealing with blasphemers! He said the name of God is regularly invoked to do the Devil’s work and to beware of those who must publicly proclaim their piety, for true piety is a private thing.)

Jim fared only a bit better during this period with an accuracy rate of 4.9%. His science has a certain appeal and seems to be a logically consistent system, but Gulliver in his travels encountered similar closed systems of thought that were equally dubious. Big-endian or Little-endian, it’s still only an egg. (Unfortunately he too suffers from rationalized, self-aggrandizing accuracy assessments.)

Ev. For a while there was a fellow named Ev making predictions on Syzygy and it sounded as if he was onto something. My Dad also told me about Professor Irwin Corey. My Dad thought he was dead. It is my guess he merely changed his name to Ev. Like Corey he mixed in a small dollop of sanity with a huge load of crap, but he did it so smoothly it took people a while to notice. Like Corey, he seems to have disappeared.

Charles had 0% accuracy. I’m not sure which God it is he keeps hearing from, but maybe it is Loki.

Ray G. had the highest accuracy rating, but made so limited a number of predictions that it is hard to extrapolate a consistent prediction rate. Were it not for the Hector Mine quake and its aftermath his accuracy would have been negligible, but his predictions did follow established ideas about precursor ground effects and in that small arena could be considered meaningful. Good work, Ray G.

My Statistics professor calculated what a virtually random predictor would score (assuming the predictions were primarily made in the more active zones, especially the Pacific Rim.) He calculated a random success rate of 5.12%. I was shocked that this implied that throwing a dart at a Pacific Rim map would produce more accurate predictions than the localized body pains, planetary interactions, emails from Jesus, or other sources of precursory data usually spewed on the boards. Ultimately, in my young and humble opinion, this is all so very pathetically sad.

In conclusion, I hope those who visit these sites will learn to enjoy them for their amusement value not any predictive value. Prognosticators from Nostradamus to Jim Jones can be amusing if looked at with a tilt of the head and a bottle of beer. I suggest that folks do a little clearer thinking about where and when quakes occur and then make some predictions of their own, without the help of false gods, cat vomit, feet twitching, moons wobbling, or any other pseudo-scientific bases.

Peace and Love,

Jason.