Make More Money by Tuning Your Automated Forex System

Are you one of these types of forex trader?

- You just purchased your first forex robot, and you’re ready to jump in and start making some big money
- Your forex robot has been making great money for while but profits have decreased.

If this is you, now may be the right time to tune your forex robot.

Backtesting and tuning is essential before you first trade a system. The system you purchased may work great ‘out of the box’, but it is always wise to test the system before trading real money.

Knowing how and when to tune your system is equally important. No system will be profitable forever. Forex market conditions change possibly making your winning system a loser. However, backtesting and tuning can get your trading system back in line with the market and making money again.

Historical Backtesting

Backtesting is running a system against historical market data to see if the system may be profitable. Almost every forex trading system marketed in trading magazines or on the Internet shows backtest results. Here are some important statistics to look for:

• Profit factor – This is the ratio of total dollars (or other currency) on winning trades to total dollars on losing trades. A profit factor above 1 means that the system was profitable in backtesting. In my experience, trading systems with a profit factor below 2 in backtesting may be unprofitable in live trading. While a higher profit factor is better, a profit factor above 5 indicates that the system is overly optimized for historical data and may be unprofitable in live trading.
• Win/Loss Ratio – This is the ratio of winning trades to losing trades. There is no ideal number for this statistic, so you should pick a system that fits your personality. For example, if you have a low tolerance for losing trades, pick a system with a higher win/loss ratio. Likewise, if you can handle more losing trades, a system with a lower win/loss ratio is okay. A trading system with a lower ratio must have a higher average winning trade compared to the average losing trade (see below).
• Average winning trade to average losing trade – For any system, you will want to know the amount of your average winning trade compared to the amount of your average losing trade. Like win/loss ratio, there is no ideal number. This statistic works in conjunction with win/loss ratio. The higher your average win to average loss, the lower the win/loss ratio can be and vice versa.

It is also important to know the period of the historical data used for the test. Ideally, the system should be tested for as much historical data as possible. The historical period should also end as close to the present as possible. This approach tests the trading system under a wide range of market conditions.

Walk Forward Optimization

Walk forward optimization is a similar but alternative approach to testing trading systems. Walk forward optimization uses historical data, but adjusts the trading system’s parameters during the backtest. For example, the parameters may be set using three months of historical data then tested on the next month to get the trading results.

So, you might test and optimize the trading system’s parameters from September 2007 to November 2007 then use those parameters to run the system on December 2007 data. This process is then ‘walked forward’ to the present. Trading systems that are sensitive to current market conditions use walk forward backtesting and optimization to see more accurately, how the system would perform. Walk forward optimization is more time consuming than simple backtesting but may identify more profitable system settings.

Forex System Tuning

Trading systems that are fully backtested and making money will eventually be out of sync with market conditions. When this happens, profits will go down, and the system may become unprofitable. Trading profits can be restored by backtesting and tuning the system for the current market conditions. Ideally, you would use both backtesting methods above to determine the most profitable settings for the trading system. Tuning your forex trading system can significantly increase the profitability and extend the useful life of the system

Conclusion

You should backtest and tune a forex system in the following situations:

• Before trading a new system live (with real money)
• When a profitable system becomes less profitable or starts losing money

Backtesting and tuning the trading system will make sure that the system has the most profitable settings for the current market conditions and the currency pair you are trading.

I will post an additional article in the future that describes the backtesting tuning process in detail.