Earlier in this blog I asked you about mechanical systems, and you said that mechanical systems work in this markets. But right now i am reading a named The Intelligent Investor- Benjamin Graham. And in that it is written that All mechanical formulas for earning higher stock performance are “a kind of self-destructive process—akin to the law of diminishing returns.” So, how will you justify this. Please reply.
A mechanical system is a set of rules that can be backtested. For example: A four week breakout rule is: Buy when th Nifty closes at the highest of the last four weeks close. Exit the long position when the Nifty closes below the last four weeks lowest close. This rule is then tested over the past 8 years. It makes money. Then, it is taken as a mechanical system, because the results could be proved over past data.
This is not so bad, is it?
What is the alternative?
Hunch? Mr X says "I feel abc is going to go up. You should buy it". Now, there is no way of knowing if such calls will make money consistently. One call may be a multibagger - the reason could be luck. But unless the methodology is not systematic and tested, there is no way of saying that trading calls coming from Mr X will finally make money.
Chart Patterns? Much better than Hunch. These patterns could and are converted into mechanical systems. Rules are well defined, and, a lot of patterns can be tested with such rules. When you trade patterns based on systematic rules you are really a mechanical trader. That sounds fine, doesnt it?
What can go wrong in mechnical systems? When rules are created, based on wrong principles. But this applies to all methods of analysis - fundamental, technical, .... . So, the problem may lie in execution, but not in the basic principles.
Now I will ask a question.
If you do not know that your trading ideas have performed profitably in the past, can you trust the ideas with your money? It is possible that you can rely on third party research if you trust them, but there must be some basis. Reader feedback is requested.