Overview

This strategy is based on the famous turtle trading system and tries to follow the original rules as much as possible. It is a trend following system that generates entry and exit signals using double moving averages.

Trading Logic

• N1-day high (default 20 days) and N2-day high (default 55 days) are used to build double moving averages.
• N3-day low (default 10 days) and N4-day low (default 20 days) are used to build double moving averages.
• Go long when close price exceeds N2-day high. Close position when close price falls below N4-day low.
• Pyramid up to 5 additional long positions, each 1 x ATR (default 1) above previous entry price.
• Set a fixed stop loss at N x ATR (default 2) below entry price.
• Only allow new entry after previous trade is a winner.

Advantage Analysis

The advantages of this strategy:

• Follows trend trading principle and captures mid-to-long term trends.
• Double MAs form filters, avoiding excessive trading during consolidation.
• The stop loss setting is reasonable, avoiding too wide or too narrow.
• Parameters are configurable for adjusting risk-reward profile.
• Allow pyramiding for more profit during strong trends.

Risk Analysis

There are also some risks:

• Unable to exit timely when trend reverses, leading to large losses.
• Too much pyramiding increases trading frequency.
• Improper parameter tuning makes the system too aggressive or too conservative.
• Backtest bias, real performance may underperform.

Risks can be reduced by:

• Adding reversal signals like MACD divergence to cut losses.
• Robust parameter optimization.
• Adding position sizing to lower position size after big losses.

Improvement Areas

The strategy can be improved in the following ways:

• Add short trade logic to profit from falling prices.
• Add stop loss optimization to adjust stops based on price action.
• Add position sizing module to optimize pyramid size.
• Incorporate trend strength index like ADX to avoid false signals.
• Optimize parameters for smoother returns.
• Consider real trading costs like slippage and commissions.

Conclusion

The strategy profits by following the trend and has good backtest results. But real performance needs to be validated. Further optimization on parameter robustness, stop loss and position sizing is needed before applying it in live trading. Overall it has sound logic and much potential for improvement.