Value investors exhibit Gambler's Fallacy

Peter Bernstein continues to raise interesting points in Against the Gods.

He writes about the concept of regression to the mean, and how this has worked as a system of investing for people such as Warren Buffett and other value investors. Although he doesn't mention it, it also explain the Dogs of the Dow technique (before it became well known.)

Not to mention the "magazine cover" effect. This is where people who make the covers of magazines in their fields for their accomplishments often go downhill afterward. SPORTS ILLUSTRATED has actually studied it, because there're numerous examples of players and teams who have never been so good as they were before making the cover of that magazine. One options guru I know of (you likely get mail order advertisements from him) uses magazine covers as a contrary indicator in his system. If a company or CEO makes the cover of a business magazine, it's time to dump the stock!

It also explain the Growth Trap as explained by Jeremy Siegel in his latest book The Future for Investors. When you pay a premium for a stock with great growth prospects, you don't do as well as you do when you put the same money into an investment that is not comparatively overpriced. He compares long term investments in IBM and Standard Oil, and in China and Brazil. (Yes, in the past Brazil has been the better long term investment despite all its political and economic problems and despite China's phenomenal growth -- keep that in mind the next time you get sales letters in the mail telling you that you're going to be poor if you don't invest in China and India.

However, keep in mind Bernstein's caution that it's not always easy to know where the mean is and when the regression will occur. People who invested in the stock market in 1930 because they thought the crash was over lost a lot of money.

In gambling circles, the belief that random results will soon revert to normal because they've been abnormal, is known as the Gambler's Fallacy.

A lot of people have lost a lot of money to casinos because they were sure the roulette's wheel long streak of red "had" to end on the next spin, or 7 had to come up, or the slot machine had to pay off . . .

regression to the mean
regression to the mean