Fractal-TSI Session Breakout System
Family: breakout · Regime: trending · Complexity: high · Asset classes: Forex, Equities, Indices · Timeframes: M15, H1
Thesis
Market trends are often initiated by a breakout from an early-session equilibrium. By using Heikin-Ashi to confirm the trend regime and TSI_CD to confirm momentum acceleration, we filter out false breakouts. The Open Source Fractal Model provides the 'structural logic' (identifying liquidity sweeps), ensuring that the breakout is not just a volatility spike but a shift in the market's underlying sub-structure. We assume an edge exists in following the path of least resistance identified by these three distinct analytical layers.
Components
- Heikin-Ashi Candles (regime) — Filters out minor price fluctuations to identify a sustained trend regime, ensuring trades are only taken in the direction of the smoothed trend.
- TSI Convergence Divergence (TSI_CD) (direction) — Measures momentum acceleration via the histogram color change, providing a filter to ensure the breakout has active momentum support.
- Brackets Indicator (entry) — Establishes the session range. A breakout of these levels serves as the primary liquidity-based entry signal.
- Momentum Oscillator (exit) — Identifies the exhaustion of the rate-of-change; an exit is triggered when the asset's relative speed begins to mean-revert.
- Open Source Fractal Model (risk) — Provides market structure context. Uses 'Change in Sub-structure' (CISD) and Fractal levels for dynamic stop-loss placement and risk quantification.
Known failure conditions
- Price enters a prolonged tight range within the bracket levels without a decisive breakout.
- The Fractal Model fails to generate a CISD level within 10 bars of a bracket breakout.
- High-frequency flipping of TSI_CD histogram colors during the breakout phase, indicating 'wash-out' volatility.
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.