KNN-Laguerre Hybrid Trend Scalper
Family: pullback · Regime: trending · Complexity: high · Asset classes: Equities, FX, Crypto · Timeframes: H1, H4
Thesis
Market trends are most tradable when institutional volume (VWMA) aligns with historical price patterns (KNN). By entering on volatility-defined pullbacks (Bollinger) confirmed by low-lag momentum (Laguerre), we can capture the resumption of a trend before significant momentum is lost, exiting when selling pressure (Bears Power) begins to decouple from the price trend.
Components
- Volume SuperTrend AI (Expo) (regime) — Defines the high-probability trend regime using volume-weighted volatility and KNN pattern matching to ensure entries align with institutional flow.
- Ichimoku Kinko Hyo (direction) — Acts as a secondary trend filter; price must be on the correct side of the Kumo cloud to confirm directional bias.
- Bollinger bands / Connectable [Azullian] (entry) — Triggers entries on mean-reversion pullbacks (touching/crossing bands) within the established trend.
- Bears Power (exit) — Signals exit when momentum wanes; specifically when selling pressure returns in a long or buying pressure in a short.
- VWAP + EMA Cross Pullback (risk) — Provides dynamic ATR-based stops and a fixed reward-to-risk ratio for risk management.
- Laguerre RSI (confirmation) — Provides low-lag momentum confirmation to filter out premature entries during choppy consolidation.
- Gator Oscillator (volatility_filter) — Ensures the market is in an 'Awakening' or 'Eating' phase (expanding volatility) to avoid low-liquidity traps.
Known failure conditions
- Strategy underperforms in low-volatility 'sleeping' Gator regimes for extended periods.
- KNN classification consistency drops below 50% accuracy on historical lookbacks.
- Price repeatedly crosses VWAP/EMA without sustained trend development (whipsaw).
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).