Trading via technical analysisÂ is mainly an effort in pattern recognition. Evaluating the price action on a chart is pattern recognition. Using oscillator type indicators such as Stochastics are about pattern recognition. Even working with simple moving average crossover methods are pattern recongition efforts. In all cases it’s about seeing something which is showing up now and extrapolating from that a forward looking expectation.
Of course, before we can develop those expectations we have to be able to identify the patterns.
When we initiallyÂ learn a chart pattern or an indicator, part of that education is naturally going to include some basic rules of thumb for pattern recognition and application thereof. Here’s the problem, though. While the patterns may appear in a consistant way – such as RSI rising above 70 – the implications of them are rarely the same twice.
Here’s an example. Keeping with RSI, a reading over 70 is generally the point at which one starts to think of the market as being in overbought territory. Trading for a reversal based on that reading, however, only makes sense in a primarily range-bound market. In a strong trend, however, it’s definitely not a good idea to do that. It’s a question of understanding the tool – knowing the patterns.
Here’s the thing, though. Real pattern recognition can only come from experience (and knowing in detail how the indicator is calculated, if using one). That means getting deep into things – spending a lot of time watching price movement and how the patterns develop – is a must. Much of this will come from real-time application, of course, but things can be accelerated by back-testing.
By back-testing here I’m not really talking about system evalution. I’m using the term to describe a process by which you go back and walk through the historical data as if you were trading using the methodology (indicator, chart pattern, etc.) in question. This means some grunt work, though. You have to really do a period-by-period walk-through, not some automated run. It’s about you learning to see the patterns, not about your computer performing some pre-programmed action.
I’m not going to say doing this type of back-testing is the same as having live experience. It’s not. What it can do is move you well down the path toward mastery, though. One of the major issues which leads to new trader failure is that they never spend enough time with a method to really understand it and to learn the patterns by which it can be successfully employeed. They instead bounce from one to another, always looking for the easy solution. There are none. Get your hands dirty. 🙂