KNN-FVG Structural Imbalance Trader
Family: hybrid · Regime: trending · Complexity: high · Asset classes: FX, Equities, Indices · Timeframes: H1, H4
Thesis
Institutional 'Smart Money' footprints (FVGs) provide directional bias, but entries are only valid when confirmed by machine-learning-filtered volume momentum (KNN) and when the market structure is not yet in an exhausted climax state. By using a point-based grid for risk, we normalize volatility-induced noise, targeting structural market shifts.
Components
- Combined Candle Counter (CCC) Dev (regime) — Provides the high-level structural regime, identifying distribution days and exhaustion climaxes to prevent entering at the end of a move.
- ICT + FVG + RSI V1.2.12 (direction) — Identifies institutional liquidity voids (Fair Value Gaps) and momentum extremes to determine the directional bias.
- Volume SuperTrend AI (Expo) (entry) — Uses a KNN-refined volume-weighted trend to trigger entries once volume and momentum align with the HTF bias.
- OTLIB Stochastic Oscillator (exit) — Provides cycle-based exit signals when the local momentum reaches an exhausted state (overbought/oversold).
- Point-Based Price Grid (risk) — Provides a fixed-point structural anchor for stop-loss placement and position sizing, ensuring consistency across different price levels.
- ATR Heiken Ashi (volatility_filter) — Filters out low-volatility periods where FVG signals are likely to be noise rather than institutional intent.
Known failure conditions
- The 'Point-Based Price Grid' fails to adapt to high-volatility spikes, leading to stops that are too tight for the regime.
- Distribution Day logic in CCC fails to detect institutional selling due to the 0.998 threshold being too rigid for high-beta assets.
- KNN classification clusters around 'neutral' noise in ranging markets, leading to rapid signal flipping.
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.