EMA Trend Filters with Donchian Breakouts and ATR Risk Sizing

Summary

This trend-following system combines a fast and slow exponential moving average with a Donchian Channel breakout. It enters long when price closes above the prior channel high while the fast average is above the slow average, and enters short below the channel low when the averages indicate a downtrend. The description gives 40 and 120 periods for the averages and a 50-period channel. Position quantity is derived from a stated fraction of initial capital divided by an ATR-based risk distance; the source’s exits use a four-ATR stop, despite the parameter list describing a different multiplier.

The material explains the rationale for filtering breakouts with trend direction and using volatility-scaled risk, and discusses whipsaws in sideways markets, false breaks, and parameter sensitivity. It supplies no performance evidence beyond a brief Bitcoin futures backtest configuration, so its claims of stable returns are unsupported. The source also includes trailing-exit behavior whose implementation and relationship to the stated stop are not fully explained.

Key ideas

• The system uses fast and slow exponential averages to filter the direction of channel breakouts.
• Donchian channel boundaries are based on earlier highs and lows, so the current bar does not set its own breakout threshold.
• Position quantity scales with ATR and a stated risk fraction of initial capital.
• The source uses a four-ATR stop even though the parameter list specifies another multiplier.
• Sideways conditions, false breakouts, and market-dependent settings can undermine results.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.