Quote from Jreality:
I agree that his "Simple Strategy" (his normal trend following strategy) isn't too profitable unless you happen know when NOT to take the trades and that ain't easy to know.
Alright, look. I'm going to tell you something that I probably shouldn't, but this is the key to the entire markets. If you can figure this out, you will make a killing. Most people don't actually realize this, although all successful traders probably do.
Think about this question:
If I told you that tomorrow was going to be a heavily trending day for 100% certain - do you think you could make money? Of course you could. A five year old could make money. All you would have to do is buy/short an MA crossover, and get out when it crossed back. Simple as pie.
Now if I told you that tomorrow was going to be a range-bound choppy day, for 100% certain - could you make money? Of course you could. Just use any oscillator you like, buy when it turns at the bottom and short when it turns at the top. Alternately, buy support and sell resistance all day. You would make a killing.
So you see, you already have all the methodologies you need to make all the money in the world. The trick is knowing when to use each system. There are some indicators like ADX which are supposed to detect trending conditions, but they are generally complete crap so don't look for some magic indicator to tell you. Personally I have subdivided the market into 6 separate condition types, each with its own separate trading system designed to extract the most amount of money during the time in question. (I know to the tick exactly how far retracements are likely to go in each condition type, for example.) Trading experience and intuition is what you need to seamlessly transition from one mental state to another as the market changes all day - and this is what it takes time to learn.
The bottom line is that you have to use your human intuition and experience to detect trend vs. chop, and then apply a mechanical system once you have determined the correct condition. A computer can't determine trend vs. chop very well - your mind is a lot better at it. As I said, there are clues for what to look at to figure this out with a high degree of accuracy, but I'm not going to tell you any more than I have already because it's getting close to revealing something I don't want to.
What you really need to do is develop a separate trading strategy for each market condition (something that should literally take about 5 minutes), and then spend all the rest of your time and effort thinking about how to detect when and how the market is transitioning between range-bound and trending states. If you can do this, you will make money all the time.
Forget about running one system in all market conditions, that's stupid - only idiots do that.
This is all I can tell you.
