Its funny how Albert Einstein is widely considered to be a paragon of moral wisdom, just because he discovered the theory of relativity and turned Newtonian physics on his head. Does an understanding that the speed of light, mass, and energy are all interconnected make you a better person? Sure, Einstein’s relative pacifism, very progressive views for his time and abhorrence of nationalism in the build-up to the second world war, as well as his status as a refugee from Nazi Germany, are deeper reasons why he became a political pin-up.
Einstein showed that it doesn’t matter where you were or how fast you are moving – the speed of light always appears to be equal to every observer. I thought about this earlier today when I tested a trading system with five different sizes of trailing stops, and found that all were profitable, and that there was not a great deal of difference between them. Provided one constant was selected and stuck with, it was profitable over the long run. The danger would be in changing it from trade to trade. It is the principle of the trailing stop which is the speed of light here, not the precise value of the stop used.
Very often you will read about trading systems which tell you to use a twenty period EMA and a fifty period EMA, for example, but the precise numbers matter little. This is what too many traders can’t get into their heads. Building a good trading system isn’t about finding the exact numbers which worked best in the past. Building a good trading system is about finding PRINCIPLES which work the same across a range of parameters.
A principle might be, buy when the price is making new highs. Let’s say you find that buying at 2-month highs does not work very well but buying at 3-month highs does. Does that mean 3-month highs are the way to go? No. What you should do is test what happens if you buy at 1-month highs, 2-month highs, 3-month highs, etc. Are they all profitable, or mostly profitable? If so, then the principle of buying at new highs is a solid one.
What I have been saying in plain English can be translated into trader language: don’t over-optimize your trading strategies based upon back-tests. Another powerful tool is back testing with a few years of historical data, and then performing a mock forward test on a separate period of “out of sample” historical data. If the result is good, you are likely to be onto a winner.
These two tools can save traders using mechanical trading strategies a lot of money and heartache.