Tongue firmly planted in cheek, I described a mechanical trading strategy that was impressive in its ability to lose money. The logic behind the strategy was based upon the stock market's tendency to reverse swings of strength and weakness. I believe this logic has worked because we have a great deal of money chasing a limited number of opportunities in the equity world. This creates increased noise and volatility at all time frames.
I don't find such choppy conditions limited to U.S. stocks. Let's take emerging markets equities, as represented by the EEM exchange-traded fund. We can create a 20-day moving average of EEM and Bollinger Bands around that average at a width of two standard deviations. Note that two standard deviations is a metric that scientists employ to determine whether or not the difference between two means is statistically significant. So, in a loose sense, we can create a trading system that waits for a "significant" trend before jumping aboard. If EEM closes above its upper Bollinger Band, we would gauge this as a signficant uptrend and would buy. If EEM closes below its lower Bollinger Band, this would qualify as a significant downtrend, and we would (short) sell. Positions in both cases would be closed out on the first daily close within the envelope defined by the Bollinger Bands.
Over the last three years, such a system would have generated 46 trades in EEM. Eleven of those trades would have been profitable and 35 would have lost money--another system that loses three-quarters of the time. The average profit per winning trade was .48, and the average loss per losing trade was .79. Overall, across the three years, the system lost an incredible 22.65 points (EEM began the period trading near $33 per share).
It takes real skill to find a system that not only loses more often than it wins, but also produces larger average losers than winners. But that is what has happened when traders have waited to jump aboard trends once they have become "significant". Once a trend has been clear, a reversal has been near--abroad, as well as here.