Don’t pull out, “you don’t understand, it’s all about statistics.” I will pull a Who’s Who PhD of statistics out of my back pocket and let him argue with you. This Gambler’s Fallacy drives him wild. The idea of being able to predict random events by looking at previous random events is very popular with gamblers and gambling investors. The idea is as popular as it is unsuccessful. Famously, in 1913, gamblers saw the roulette wheel come up black 26 times in a row and bet against black. The wheel, not having been informed that it “owed” anyone red, came up black one more time, and all the gamblers lost. This kind of failure has continued uninterrupted, but despite over 100 years more evidence, gamblers still believe the Gambler’s Fallacy, to the delight of the house.
What does this have to do with investing? Many investors assume that they have figured out the pattern for a stock. They carefully analyze the ups and downs, see a pattern, and then bet that they’re right. If they’re right, they say it proves their method works. If they’re wrong, they point out that no system is 100%. Fair enough. Let’s see what the studies say:
- 2004 in Taiwan: 80% unprofitable, 19% break even, 1% profitable
- 2012 in U.S.: 70% unprofitable
If people guessed at random you’d expect to see about 50/50 profitable and unprofitable, skewed toward unprofitable because of transaction costs.
What’s the bottom line? You’re better off getting rid of your system and flipping a coin. The results will probably be a lot better. Better yet, if your goal is to make money, buy and hold a low cost index fund.
