The question
Over the ten years we tested, most big US companies went up. In a market like that, almost any way of picking stocks makes money. So a good-looking result does not show that a rule is any good.
We wanted to know how much of each strategy's result came from its rule, and how much came from simply being in the market.
How we tested it
We ran the test again with no rule at all. Every trading day, went into one company picked at random from the companies that were in the S&P 500 that day.
Everything else stayed the same: the same dates (), the same costs, the same limits on one stock and one sector, buying at the next day's opening price and never selling.
We did this three times, with three different sets of random picks. One random run can be lucky or unlucky. Three give a range.
The result
What it means
What it does not prove
- It does not prove the rules will keep working. These ten years were unusually kind to large US companies whose prices were already rising, and every rule in the lab leans on that in some way. A different decade could give a different answer.
- Three random runs is a small number. More runs would give a firmer range. The original research runs, on a different set of price data, got 13.0% to 13.8% a year from their three random runs, which is close to what we see here.
- Nobody invested this money. It is a test on past prices. Live results are likely to be lower, and the live record is what counts in the end.