When you are developing an automated strategy, you’re building a set of conditions that are referred to as
signals for market
entries and exits. You then combined these conditions (trade entry, risk management and money management,) together to create a
system. In addition, you can build
filters to determine when the system should run vs. not run according to market conditions.
You then begin
back testing your strategy you built on one timeframe to see how it has performed from a historical perspective. Some traders also run their newly developed strategies going forward in a test mode (but don't take the trades) to see how it performs in real-time (sort of like a prototype). This is referred to as
test trading and then they eventually move the system to
live trading.
So you want to have some experience with your system either in test mode and/or historical (back tested) to see if it has a positive expectancy. Traders who try to build a strategy that accommodates multiple market conditions (bullish, bearish, and neutral) have a difficult time integrating this with the conditions of the security! You have to ask yourself which is more important, the market conditions or the conditions of the security I am trading?
This is a tricky question. That's why some traders will voice an opinion of systems not working. You have to learn to build filters to assist the strategy on the security to address the market conditions separately.
For example, when the e-mini's (proxy for overall market) are all trending upward, and your strategy gives a signal to short, do you want your system to take that trade? You might say, probably not, since the odds are not as good and in your favor. It still might be a great trade, but are you trying to trade with the market or against it! Even if your back testing suggest otherwise, you have to determine is the trend your friend or trade against the trend.