Nobody knows nothin’ about crude oil prices.

Today's weekly Bloomberg survey of oil analysts and traders projects that prices will rise during the week of August 25. The story reports that the survey, which has been done weekly since April 2004 has correctly predicted the direction of market movement 49% of the time. It seems Bloomberg could improve the accuracy of its prediction if it just reported each Friday the results of a coin flip. And, since the coin flip would take much less time and money than the survey, maybe Bloomberg could afford to expand its service and give us a daily prediction.

Update on Friday, August 22, 2008 at 01:05PM by Skeptic

And nobody knows how to forecast supply and demand either, according to this report.

Update on Wednesday, September 10, 2008 at 08:16AM by Skeptic

But of course financial journalists and the traders they interview always have to cite some reason for a price movement.  They can't say "oil went up today for no particular reason," or that the price decline today was apparently a random event.  The pretense at understanding is even more exposed when one considers that the same piece of news can be ascribed opposite meanings.  For example, in a rising market, a report that crude oil inventories had declined would probably be considered bullish based on the assumption that somebody was short and would have to boost his inventories and pay up to do it.  On the other hand, in a declining market, lower inventories might signal that the real insiders expect to be able to buy what they need later at a lower price.  (The linked post also refers to an updated Michael Masters analysis of the extent and effects of large reductions in crude oil futures investments by financial players over the summer, but I have not tracked down the Masters update.)