Markets: Efficient, More or Less

Kash Mansori is arguing by analogy today: if the NFL draft is inefficient, doesn't that mean that markets might be inefficient too?

If people also have blindspots when they deal on the financial markets, that would imply that markets are indeed not completely efficient in all cases. That in turn suggests the possibility that there may be publicly available information that could allow one to profit in a systematic way on the financial markets, like the Oakland A's did in Moneyball.

I think that Kash misses the point. The world's most hardline proponents of the efficient markets hypothesis don't believe for a minute that people, individually, are efficient and not prone to blindspots. What they believe is much weaker than that. They believe that if you have thousands or hundreds of thousands of market participants, then when you put all those errors together they have a tendency to cancel each other out. It's known as the wisdom of crowds.

The NBA or the NFL can easily violate the efficient markets hypothesis because the number of teams involved in the market is low, and the number of trades that are made per season is also small. If you had a few thousand teams making tens of thousands of trades per day, then I think it would be a lot harder to demonstrate that the markets were inefficient.

As for me, I think the efficient markets hypothesis is probably false on an absolute level: I don't think that markets are perfectly efficient, and I think that in fact most intraday price movements do not reflect real changes in the value of the underlying companies. On the other hand, I also believe that the probability of any given individual being able to successfully arbitrage market inefficiencies is minuscule to the point at which even trying to do so is folly. Yes, it probably has been done in the past. But the chances of you doing it – or the chances of you being able to pick a money manager who can do it – are so slim that you really shouldn't even bother.