Person A: I have a strategy. Person B: Test the strategy with 100 to 150 trades. Person A: I'm too lazy for that, I'll just use a back-testing machine. Person B: I see Person A: This sucks, back-testing gave negative results Person B: Is it possible if you actually trade the strategy you might get better results? Person A: Dude, that's so antiquated. Person B: I see Person A: I have a strategy. Person B: Test the strategy with 100 to 150 trades. Person A: Why do you keep telling me that? Person B: It might actually help you learn how to trade. Person A: Freak that, I trust software more than my own ability. Person B: I see Person A: This sucks, back-testing gave negative results Person B: Have you ever seen the movie Groundhog Day?
Agree. I tend to come up with more nuanced and interesting systems while forward testing with actual money because there is actual risk involved. Plus, there are some human decision processes that are grey areas and not quantifiable and are impossible to backtest.
Back testing for the most part is only useful in automated systems. Using PA, you can trade in real time at the hard right edge of the screen. I know I will take shit for the following statement, but many advanced traders can actually do what is called "always in" described in 100 pages deep in ET. What this means is you know the market is going down, so you trade down, then market goes sides ways or up, and you either get out or reverse your position. This creates turns, the turns are how you achieve profit not with a profit target but when for example the market reverses, you are now long instead of short and you build your profit throughout the day.
Forward testing is conceptually 100% identical to backtesting. You base your future results on the exact same premise, which is that you believe that future market dynamics will highly correlate with the market dynamics during forward testing. The data on which the testing is based is just across a different time frame. Otherwise identical. Never understood the concept of forward testing. Either you believe past dynamics somewhat repeat themselves or you don't. If you don't then any automated/quantitative approach is futile, back or forward tested.
Backtesting is a tool that only works for people that know how to trade and make money in the markets. For all the other people that do not make money trading, should stay away from backtesting. You do not become a profitable trader by finding a system that makes you money according to a backtest. Read that sentence twice. It will save you a lot of time.
You can even make it better. In the area where the market turns, there is almost always a consolidation period. In that period the market does some kind of range trading. I always get out in that consolidation period, so risk is reduced as I am out of that position. Another advantage (if you are able to do this) is that when I enter the consolidation period, I try to find extreme points which gives me more profit than the real moment when the market turns. And with some luck I can also catch an extreme good entry for my next trade. If you are however always having FOMO, this will not work good for you.
First, back test must cover the entire market. Assume no capital constraint. Use those data and focus on area of profitability.
That always-in thing is for sure not my thing. I believe I can not predict when a move against my positions is a correction or a directional change of trend. Therefore I want to protect my profits as much as possible and will get out of the slightest weakness against my position. If it turns out just to be a correction, a la , re-enter, maybe even on a better price because I did not stay in for the move against me...
Good Morning danielc1, I am confused by your statement. Can you please explain in more detail? Thank you