McGinley-SGF Poly-Volume KNN Trend

Family: hybrid · Regime: trending · Complexity: high · Asset classes: FX, Equities, Crypto · Timeframes: 15M, 1H

Thesis

The strategy assumes that price trends exhibit non-linear momentum that can be isolated using polynomial filtering (SGF) and speed-adjusted averages (McGinley). By further filtering signals through a KNN algorithm that looks for volume-confirmed trend patterns and ensuring entry occurs at institutional liquidity zones (SNAP), we gain an edge by entering only at the confluence of noise-reduced direction and structural support.

Components

• McGinley Dynamic (regime) — Acts as a 'smart' regime filter that avoids the lag of EMAs during volatility spikes, ensuring we only trade in the direction of the dominant trend speed.
• Savitzky-Golay Filter (SGF) (direction) — Filters out high-frequency market noise to determine the 'true' directional bias using local polynomial regression without the phase lag of traditional filters.
• Volume SuperTrend AI (Expo) (entry) — Uses KNN classification to validate the VWMA-based trend, ensuring entries have both volume support and historical pattern similarity.
• UTBot Alerts (exit) — Provides a volatility-sensitive trailing exit that reacts to trend exhaustion via ATR-based price crossovers.
• SNAP HTF_LTF Indicator (risk) — Identifies institutional supply/demand and liquidity pools to define logical stop-loss placements and avoid entering into overhead resistance.

Known failure conditions

• Strategy invalidates if KNN classification accuracy drops below 50% in a 200-sample backtest.
• Market enters a prolonged low-volatility 'staircase' where UTBot's tight ATR-1 setting causes constant premature exits.
• SGF order and window size produce significant aliasing on the specific timeframe used.

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.