Disliked{quote} Darren and Copernicus, Does it really matter to extensively backtest any entry system? As Darren stated several times in earlier posts, if the math suggests that asymmetrical reward/risk over a LARGE number of trades is profitable, then conceivably, ANY discretionary/technical/fundamental method that gets you to open a position should be long-term profitable IF you are methodical in executing your trailing SL/exit strategy. The best way to run such an operation, as Copernicus so elegantly described, is to diversify over many instruments...Ignored
Your logic is sound. While random entry systems do work provided you are medium to long term in your initial stop and trail settings (eg. 6ATR, &ATR, 8ATR)....from my testing, there is definitely an additional edge to be gained by ensuring short term momentum is on your side on entry and that the direction is in the direction of the primary trend. The former highlighted condition tends to get you of to a good start and the latter highlighted condition ensures that your trades can possibly run most of the length of the longer term secular trends (that can be huge).
Breakout entries with an additional momentum push help you achieve the following:
1. It can allow you to have a slightly tighter initial stop which you also use for position sizing purposes and with the Law of Large Numbers gives you a bit of leverage with regards to better risk:reward outcomes.
2. The boost of momentum can really help in the early stages of catching a trend to quickly push you away from the inevitable noise of entry.
3. To distinguish between a possible random trend from the way price collates and a real trend of substance with momentum backing it.
I would be interested to hear 60's view on this. As you know I use fairly restrictive filters to ensure that my breakouts only occur when prices get away from 'normal' conditions. For example a 50 Donchian, 100 Donchian, 200 Donchian act as different volatility filters which restrict when my systems are turned off or on. By using these filters you are effectively guaranteed that price has momentum when they are breached....but I also use additional momentum criteria just to confirm it is on the rise.
The way I tend to view trends in general is periods of consolidation and periods of momentum burst. In fact momentum is just a shorter term class of trend.
If you start off with a momentum burst....then it will help you at least lift the trailing stop to reduce any possible loss of whipsaw. It will also help psychologically by getting to blue sky more quickly where you can then forget about it and let it do it's thing.
If you set your initial stop at R and trail from the outset or when you get to BE, then the short term momentum burst will ensure your worst case loss is R (very rarely) and most less than R) which is good for your overall stats in the long term.
Cheers
C
2