Fooled by randomness…again

Posted by Marc Hodak on May 21, 2007 under Patterns without intention | Read the First Comment

The WSJ is trying its damnedest to disprove my theory that education makes a difference. In this story, tautologically headlined “In a Volatile Market, the Winning Analysts Had Their Timing Down,” they compile a list of Wall Street analysts based on the performance of shares they recommended. This is their 15th annual ranking of analysts, which means that this intellectual embarrassment was inaugurated when 20 years worth of papers had already accumulated providing empirical evidence that stock picking prowess was a largely illusory phenomenon. Yet the Journal still writes stuff like this:

Although there are many ways to rank analysts, this survey is strictly quantitative and based solely on stock picking skill.

Here is a funny definition of skill: luck.

One of the attributes of skill, the reason you know that Tiger Woods is skillful instead of lucky, is that skill results in performance persistence. A top athlete will consistently beat a mediocre athlete over any period of time. Wall Street pros, at least those with public records, don’t display this persistence. Neither analysts nor fund managers tend to stay on top from one year to the next with any greater frequency than a group that had good results picking stocks by throwing darts. Here is how the Journal interprets this phenomenon:

Several firms that performed well in previous years were tripped up by last year’s volatile market. A.G. Edwards & Sons, a unit of A.G. Edwards Inc., the No. 1 firm with nine awards a year ago, fell to No. 31 on this year’s list, with just three winning analysts. Citigroup Inc., which ranked second a year ago, sank to 49th place. Credit Suisse, a unit of Credit Suisse Group; Standard & Poor’s Corp., a division of McGraw-Hill Cos.; and J.P. Morgan Chase & Co., which rounded out the top five a year ago, all failed to place in the top 20 in this year’s survey.

Sigh.