Friday 26 August 2011

Not All Pullbacks In The Trend Are Created Equal

I trade the pullbacks in the trend of the chart I am trading. As the title of this post says, not all pullbacks are created equal. The qualitative classification of the pullback is important, particularly if you are using your algo to enter trades.


So you need patience. Patience to wait for the picture to appear that meets your TP (trading plan). But of course for that to happen, there needs to be a detailed enough trading plan. And you need to be able to recognise the picture in a timely manner and execute the trade. Quite a lot to do when the markets are moving with a lot of volatility.


But back to a qualitative assessment of pullbacks. Pullbacks are pullbacks in price, but they can be seen and measured more easily with indicators. Indicators do NOT lag. They show what they show at the moment. Using an indicator which shows you the information that you want to see is they key. Indicators show you what is happening and what is not happening.


There are people that trade who say that only price should be looked at. They say that only a bar chart should be used. But the moment you create the bar chart you are really using an indicator, as your bar chart begins organising the price into bars of your chosen periodicity. The only true price based trading are the time and sales and the bid and ask size. We had that on the floor, but we had more than that as we "felt" the market and "saw" the market, information that is very difficult to get electronically just from time and sales and the bid and ask size. So we measure the order flow using indicators, whether it's volume breakdown, CCI, moving averages and the like.


The chart below of the Euro trades today shows the two best trades out of the four. It may well be that you want to filter out the two "lesser" trades and even more "lesser trades. The way I go about this is to add my tools to the chart and observe. Once I have an idea of what will filter these lesser trades out, I back test it with a quick algo. Then I look at the behaviour of what I have created over a variety of markets and decide what else I need to do or whether there are types of markets or times of day that I want to make sure I don't trade in combination. I also decide whether to run the picture as an  algo, and whether it is more suitable for fully auto trading or for hybrid or just manual discretionary trading.



No comments:

Post a Comment