Tuesday, June 12, 2007

Trading systems and Market Dynamics

Most traders, sucessful and otherwise, have some trading system. How successful the trading system is would depend on the market dynamics. If you study the trading systems, we will find that many trading systems depend a lot on the indicators. For me, trading systems and studying of the markets and trading is all about pattern recognition. Take any indicator or any study about trading and you will see that in the background, it is only pattern recognition.

Take elliot waves for example. We try to find patterns in the chart and then base our entry and exits on how these patterns work out. We try and interpret these patterns as waves. Likewise, with fibos as well. We basically look out for highs and lows and try to find patterns and match them and use them for fibos. As these depend a lot on the patterns, which are in turn dependent on subjectivity, we find that many people interpret these waves and patterns in different ways and this is the reason why , though many people might follow elliot and fibos, their results from this vary a lot.

Trading systems which depend on indicators also depend on pattern matching. Just throw in 2 or 3 indicators into the chart and watch it. Say throw in 12 ema, 20 sma and 10 macd or whatever. All that you have to do is look for patterns. If 70% of the time, you find that when 12 ema crosses 20 ema, the price goes up, then you have a trading system. So all that you need to do is throw in some indicators, qweak the parameters and watch for patterns. If you are able to find a pattern by which price rises or falls 70% of the time when the pattern occurs, then you have a good trading system. Sometimes, the trading system may not be so straightforward. You need to consider news, whipsaws and all that.

So now when you have a trading system which works 70% of the time, are you settled for life? No...Why? This is due to the market dynamics. The market is never constant. It keeps changing its character. So what might work today may not work tomorrow. One of the major market dynamics is the correlation between pairs. Lot of people would tell you that EURUSD and USDCHF are highly correlated. When one rises, the other goes down. But when i started out trading, the correlation between these 2 pairs was huge. If one pair went up, you could bet your house that the other would go down and you could also pinpoint how much it would drop down by a great degree. For eg., if EURUSD went up by 10 pips, you could pinpoint and say that USDCHF would drop by , say 15 pips. But now, predicting the correlation between these 2 pairs is not so easy nor can it be done so accurately. Each pair has slowly started to have its own characteristics.

Likewise with the JPY crosses as well. A good example can be seen even now. For a long time, you could see that all JPY crosses moved together. If USDJPY is near its high, so would the other crosses. But you can see now that though USDJPY is near its high, EURJPY and GBPJPY are nowhere near it. This does not mean that they are not correlated. They are but the extent of correlation is becoming lesser.

Likewise, the volatility of the pairs also changes over time. Some pairs which had a daily range of 40-50 pips now have a range of 70-80 pips and the opposite is true as well.
This is just one example of changing market dynamics. So with the market changing so much over time, our trading systems also need to change and adapt to the changed market scenario. Thats why trading systems which used to work greatly a couple of years back do not work so well now. So be careful in choosing your trading system and once you have done that, keep following your systems closely to find out the time when their effectiveness becomes less and at those times, tweak it to make it more effective or move on to another trading system.

Happy Trading!!!

No comments: