080399
To many who follow these web pages, we may have given the impression that our trading focus is 100% discretionary due to the way we have used our tools to discuss trade setups. Although we do feel that discretionary trading offers more promise than systematic trading, by no means have we neglected this area of market timing. In fact, we have created many profitable systems which we have traded personally and which are now being traded by many of our older clients. The main advantages of system trading is the ease with which they can be backtested, and the clear, unambiguous nature of the trading signals.
One of our favorite longer term systems is a trend following breakout system that works very well on the currencies. We have been following it in the Japanese Yen for about 18 months now, and it has continued to perform as well as it did during the in-sample test period. It has been discussed in these pages before, but not for a long time. Below is a chart showing some recent signals:
Since it is a longer term system, the trades do tend to be widely spaced. Notice that it is also traded as a pure reversal system, which means that it is always in the market. The premise of this system is very simple. What we have done is taken a very basic breakout method and applied MA_Momentum+MA_PredictDELUXE as a timing filter. In other words, we will only take our breakout signals when MA_Predict/MA_Momentum tell us to do so.
MA_Predict working on MA_Momentum produces a very accurate and responsive indicator that has served us well both in discretionary trading and when building systems. The results of adding this tool to our original breakout system resulted in a large reduction in drawdown, an increase in accuracy, and larger profits. Below are the results of this system from 1/1/92 - 8/2/99:
These results are very good, and are competitive with any JY system in Futures Truth or the popular trading magazines. These results could easily be duplicated by anyone with the same tools and a little time on their hands for testing. Stops, profit targets, and reentry/double down signals could all be added to our basic system to improve the performance, so there is definitely room for improvement. Right now, the system averages around $12,000 per year. With some tinkering, it should be possible to raise this to around $15,000, which would be better than almost anything for sale. The best systems are always home made and closely guarded, so if you like the idea of system trading, you might want to think about trying to build your own. Stay tuned...