Confessions of a Day Trader, Part 1: How Do We Test Technical Analysis?

Greeks, elves, and flags oh my!  The efficient market theory (the mainstay of any modern business school education), teaches us to view technical analysis with extreme skepticism.  The theory goes like this: if investors predict that a stock will rise in the next week, they will immediately buy the stock and drive up the price to the expected future price (discounted for the time value of money, of course).  But is it possible to spot something before everyone else and make a billion dollars before the market catches on?  …or are day traders just modern alchemists?

Economists like George Soros and Benoît Mandelbrot have long suggested that the efficient market theory doesn’t really hold water.  They argue that, when put to the test, the stock market follows something of a pattern or, at the very least, it doesn’t follow the same behavior as a series of coin tosses.

I have come up with a test of technical analysis that was suggested to me by Barry Nalebuff and Avinash Dixit’s new book, The Art of Strategy.  Over the next few days, I’ll apply the same test to the stock market that three psychology professors applied to basketball stats to see whether the stock market is truly random.

If my experiment suggests that the market is not random, will I be in a position to make a billion dollars?  Unfortunately, no.  To make money off my strategy, it will have to pass a number of other tests:

• Was the correlation between expected and actual results significant enough to be predictive?
• Were the predicted changes large enough to represent a viable investment strategy, after accounting for taxes and transactional costs?
• Can we reproduce the experiment with real money?

Since I’m blogging about it, the answer to that last bullet will almost certainly be no.  That’s one area where I’m fairly confident with the efficient market theory…if it’s publicly known how to make money on something, the opportunity will disappear.  On the other hand, the audience for my blog is small (as in, one guy and his dog read by blog on a regular basis).  So, who knows?  Maybe the dog will get rich.