Why do some black box strategies that ‘worked,’ stop working when you use it?

May 14th, 2009

Share |

Let’s imagine you come across an online advertisement that tries to sell you a ‘secret’ trading technique that brought investors untold returns over the past, say 20 (or 30 or 40 or whatever) years. That technique is based on complex black-box algorithm that involves processing huge amount of data. The underlying message from the advertisement is that since this technique is successful over so many years, it is one that ‘works.’ That advertisement may show you past trading records and perhaps even back-tests of that technique.

Impressive isn’t it?

How was the ‘secret’ trading technique derived? Such advertisements will usually claim that the ‘soundness’ of the techniques are based on studying the market over a long time and ‘proven’ by the ‘stellar’ performance of the technique. But in reality, such ‘proofs’ are an illusion. Let’s turn to the Chapter 1 commentary of Benjamin Graham’s The Intelligent Investor:

?If you look at a large quantity of data long enough, a huge number of patterns will emerge?if only by chance. By random luck alone, the companies that produce above-average stock returns will have plenty of things in common. But unless those factors cause the stocks to outperform, they can?t be used to predict future returns.

This is the basic thesis of Nassim Nicholas Taleb’s book, Fooled by Randomness. Therefore, buyers beware!

Tags: , ,