Volatility Breakout Strategy (C#). StockSharp

Author: StockSharp
N: 1724
v5.0.1 (7/30/2025)
Downloads: 73

The Volatility Breakout strategy seeks strong directional moves when price escapes from its average range. By measuring the distance from a simple moving average using the Average True Range, the algorithm defines breakout thresholds that scale with volatility.
A buy order is triggered when the close rises above the SMA by more than Multiplier times the ATR. A sell signal appears when the close falls below the SMA by the same distance. Positions remain open until an opposite breakout occurs or a protective stop is hit.
This technique caters to intraday traders who thrive on momentum surges. Using ATR-based thresholds helps filter out noise so only significant moves generate trades.

  • Entry Criteria:

    • Long: Close > SMA + Multiplier * ATR
    • Short: Close < SMA - Multiplier * ATR

  • Long/Short: Both sides.
  • Exit Criteria:

    • Long: Exit when an opposite breakout triggers or stop-loss hits
    • Short: Exit when an opposite breakout triggers or stop-loss hits

  • Stops: Yes, stop-loss at Multiplier * ATR from entry.
  • Default Values:

    • Period = 20
    • Multiplier = 2.0m
    • CandleType = TimeSpan.FromMinutes(5)

  • Filters:

    • Category: Breakout
    • Direction: Both
    • Indicators: SMA, ATR
    • Stops: Yes
    • Complexity: Intermediate
    • Timeframe: Intraday
    • Seasonality: No
    • Neural networks: No
    • Divergence: No
    • Risk Level: Medium