This question came in over the weekend from a gentleman working on improving his trend trading system.
Hi John ,
I have a question for you I’m hoping you might be able help with . I recently started researching Donchian’s Channel Breakout System to trade Forex with. Using Donchian’s 4 week rule I have my price bands set at 20 days. I ‘m sure you are probably very familiar with it. If price closed above the upper band you enter a long position and if price closed below the lower band you enter short. I like the system because of its simplicity and it appears to work on any time-frame. My question is in order to improve the winning % of trades, I have been playing with using a filter like MACD histogram but it doesnt seem to filter out many bad trades? Do you know of an oscillator that might work to help filter bad entries to incerase the winning % of trades?
Best of trades ,
I am not a Donchian expert by any stretch of the imagination. As I understand its basic go with break-outÂ premise, though, the approach is one which is focused on trend trading. That means we need to keep two things in mind.
1) Trend trading systems have low Win %
It is the nature of systems which attempt to get on and ride trends that they tend to lose more often than they win. It’s a simple function of the markets not being in real trends most of the time. Of course the idea is that when you get a good trend it more than makes up for the losses suffered during those periods when there is no trend going on. Trend trading is probably as simple as it gets in terms of trading. The problem, however, is that riding out the inevitable drawdown periods and low Win% makes the approach something many traders struggle with on a psychological level.
2) Oscillators are range trading indicators
John asked whether I know of any oscillators which could help him filter out bad trades. The problem is that oscillators are used for range trading, not trend trading. As such, they really work at cross purposes to things like Donchian. The trend trader is going with a break out at basically the same point as RSI (for example) would call the market overbought.
The only way to improve the Win % of a trend trading system is to await further confirmation of the trend being in place by using some kind of indicator or reading that is a bit more lagged than the one you normally use. Of course that then means your gains are reduced. In the final analysis you may find it doesn’t increase the expectancy of your system at all.
Don’t fixate on Win Rate!
I’ve said it many, many times. One of the biggest mistakes traders make is getting to caught up in how often they having profitable trades. It’s only one part of the equation. You also need to factor in the size of the wins relative to the losers and the frequency with which trades are made. All of that comes together in expectancy.