Lorentzian Normality Breakout system

Family: hybrid · Regime: trending · Complexity: high · Asset classes: Equities, Forex, Crypto · Timeframes: 15m, 1h, 4h

Thesis

Market 'edges' exist when high-intensity cyclical momentum (Fisher) coincides with physical price force (Large Candle Bodies). In these windows, price behavior deviates from a normal distribution (Jarque-Bera), allowing machine learning (Lorentzian) to identify recurring clusters of profitable moves. The edge is structural: human participants and algos chase volatility breakouts, and the system attempts to capitalize on this 'herd momentum' before it reaches Stochastic exhaustion.

Components

• Size Highs and Lows 2 (regime) — Acts as the regime filter by determining if bullish or bearish 'candle force' (body size) is currently reaching new local peaks.
• Fisher-based Scalping Indicator (direction) — Provides the cyclical directional bias by transforming price position into a Gaussian-like distribution to isolate trends.
• Lorentzian Classification (entry) — Uses k-NN in a manifold space to trigger entries based on historical similarity to the current feature set.
• Stochastic Oscillator (exit) — Identifies momentum exhaustion for exits before a mean-reversion move invalidates the trend.
• Jarque-Bera Test (JB) (risk) — Measures the 'fatness' of tails; used to adjust position sizing based on how far market returns deviate from a normal distribution.
• Keltner Channels (volatility_filter) — Ensures entry occurs during a volatility breakout, preventing entries during low-liquidity ranging periods.

Known failure conditions

• Jarque-Bera values remain consistently above 10.0 for extended periods, indicating a permanent shift to a high-kurtosis regime where k-NN similarity fails.
• Fisher-based Scalping Indicator flattens near zero, causing rapid directional flipping.
• Price action exhibits high 'pin bar' frequency where candle bodies (SHL2) are small but ranges are large.

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.