About the "Recommended Changes" to the Trading System

What are the changes we are testing you may ask? I have recommended to the team we make the following changes to the system. Until 2 years worth of backtesting and 30 days worth of live forward testing has occurred, we do not personally feel comfortable trading large sums of our own personal money with these changes. However, I believe these changes will be enough to make the system profitable again. I also believe, these changes have the potential of being longer lasting than the previous system rules. Here are the recommendations I made to the team:

1) Automate the trading
As many of you know, we have been already working on a computer program to automatically make the trades for the system. This would take the human error component out of the daily trading. Our head trader would continue to monitor the trading system at all times but would allow the computer program to make exact entries, exits, and position size calculations on it's own. The program would allow for adjustments to risk size, currency pair, and allowed to be placed on "no trade" status as the head trader deemed necessary. This would allow the head trader to make daily assessments of the market and adjust the trading triggers to meet the needs of that day.

2) Spread the risk. Include more currency pairs at lower risk per trade
The market has changed in a way that causes our losing periods to last much longer than they did in 2007. These losing streaks are eating up all of the profits gained from our winning periods. I believe adding additional currency pairs to our Intra-day strategy would help make our results more consistent. We will start by using all 16 major pairs. Rather than taking a 1% risk per trade as we did when only the GBP/JPY and USD/JPY were being traded, we will take a 1/16% risk per trade but take 16 times as many trades. The timeframe from which each individual pair will be traded will have to be determined pair by pair. We will start by backtesting each pair traded off a one-minute(M1) chart and go from there. Without the automated computer trading system, making so many trades in such a short period of time would never have been possible.

3) Change the GBP/USD risk to reward
In 2007, the GBP/USD strategy had an over 90% daily win percentage. However, on average, it would only make 1.7% per win and 6.5% per loss. That meant that it took 4 wins to make up one loss. While the strategy has been shown to still be profitable, the emotional effects of the losing day are difficult on the fund as a whole. I have recommended to the team that we decrease the overall risk on the strategy while increasing the take-profit trigger. This will mean a lower daily win percentage but a higher reward compared to the risk. It is unknown what the percentages will be at this time but we will work with those numbers to come up with a ratio that maximizes profits while diminishing the large drawdowns. The goal is to accomplish a better than 3:2 risk to reward ratio (compared to the current 6.5:1.7) while maintaining at least a 75% daily win percentage.

UPDATE(June 23, 2008): Considering adding an additional third-party strategy to the mix. A EUR/USD Strategy.