Fractal-LSTM Hybrid Regime System

Family: hybrid · Regime: trending · Complexity: high · Asset classes: FX, Equities, Crypto · Timeframes: 1H, 4H, 1D

Thesis

Market price discovery often happens in 'bursts' that follow structural breakouts (Fractals). By using an LSTM to forecast the underlying volatility regime and an ADX to confirm trend presence, we can filter out the high-frequency noise that typically plagues breakout systems. The edge is generated by waiting for high-confluence moments where AI-optimized direction, momentum, and price structure all align simultaneously, then using a speed-adaptive exit (McGinley) to capture the trend duration.

Components

• ADX / Connectable [Azullian] (regime) — Filters for active trending regimes, preventing the ML and Fractal components from firing in low-momentum noise.
• ML SuperTrend (Ultimate) - Auto-Optimized AI with LSTM (direction) — Provides the primary directional bias using machine learning to adapt to local volatility structures.
• Trend Break Fractal (entry) — Acts as the tactical trigger, requiring a price-action confirmation of market structure violation.
• McGinley Dynamic (exit) — Serves as a dynamic trailing exit that adjusts its sensitivity based on price speed, reducing premature exits in accelerating trends.
• Grid Bot [Grid range plugin] Two Moving Avarages [psyll] (risk) — Provides volatility-derived boundaries used for non-linear stop-loss placement and position sizing.
• Average Force (confirmation) — Confirms that the breakout is backed by positive momentum relative to the lookback range.

Known failure conditions

• Market enters a prolonged tight 'squeeze' where ADX stays below 20 but SuperTrend frequently flips due to ML overfitting.
• Price gaps significantly beyond the McGinley Dynamic level during overnight sessions.
• ML model converges on a local minimum that fails to adapt to a structural change in volatility (e.g., central bank intervention).

Explore the full interactive blueprint, parameter ranges and evidence on WOBR StrategyVerse, or generate this strategy as an MT4/MT5 Expert Advisor with QuantMogul AI Engine.