There is, of course, a bit of a puzzle in finance. To wit: how can markets be so efficient that even very smart people with PhDs in finance cannot (on average) outperform a diversified index, when so many of the people buying and selling the stocks are so stupid that they think they can beat the market?

For the same reason that you can't beat the football spreads. Even if people are stupid, as long as their error is random, the errors will, on average, cancel out.

Of course, the argument of those who think that they can beat the market is that the error isn't random, and that they can therefore make money.

There are several problems with this. The first is, of course, that a lot of people have thought they discovered non-random error in the market's thinking, only to find that they were the ones who had made a big mistake.

The second is that when there are errors or information asymmetries, they tend to attract a lot of people trying to make money off of them. They rapidly bid down the arbitrage opportunity to zero.

The third is that even if you have identified a price anomaly, you may not be able to act on that information. I am acquainted with someone who shorted the stock market in 2000 and made a killing. Unfortunately, he also shorted it in 1997, 1998, and 1999, and was very close to being totally bankrupt before the market went south. As traders like to put it, "the market can stay irrational longer than you can stay solvent".

It is a common fallacy among quasi-financially-literate laymen to believe that "efficient markets theory" means that there is some platonic ideal of a price, which markets inevitably find. That is indeed a silly theory, and one easily refutable by any of dozens of kinds of evidence. But that is not EMH. EMH only says that, whatever that platonic ideal of a price is, trading on your estimate of that ideal price is unlikely to make you more money than you would by purchasing a broad market index.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.