Lorentzian Ichimoku Confluence System

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

Thesis

Market trends exhibit localized self-similarity in feature space (momentum/volatility) that can be identified using Lorentzian Manifolds. By entering only when these ML-predicted similarities align with traditional trend-following structures (Ichimoku) and momentum triggers (MACD), we can isolate high-probability trend legs. The trade is maintained as long as a confluence of 10 disparate technical layers (IAE) remains above a baseline threshold.

Components

• Lorentzian Classification (regime) — Acts as the primary regime filter, using non-Euclidean distance to identify if current price action resembles historical winning momentum profiles.
• Ichimoku Kinko Hyo (direction) — Provides a structural trend filter; price must be on the 'correct' side of the cloud and Kijun-sen to ensure the ML prediction aligns with macro inertia.
• MACD (entry) — The execution trigger that ensures entry occurs at the moment of momentum acceleration rather than just structural bias.
• IAE — Technical Confluence Score (exit) — Aggregates 10 layers of technical data; used as an exhaustion exit when the composite confluence drops below a threshold, signaling the trend's 'energy' is spent.
• Williams Fractals (risk) — Used for objective stop-loss placement at recent swing points; the 2-bar lag is accepted to ensure structural validity of the pivot.

Known failure conditions

• Lorentzian Classification 'wins' drop below 50% over a 200-trade rolling window.
• The IAE Score remains in a neutral zone (40-60) for extended periods while price trends, indicating the confluence engine is desensitized.

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