ATOM Recursive Logit SMC Expansion Strategy

Family: hybrid · Regime: trending · Complexity: high · Asset classes: Crypto (ATOMUSDT.P) · Timeframes: 15m

Thesis

The hypothesis is that ATOMUSDT 15m price action follows identifiable structural patterns (SMC) that are frequently obscured by noise. By using a Recursive Median Filter to identify the 'true' trend and a Logistic Regression model to verify momentum probability, we can isolate high-conviction expansion moves out of consolidation boxes. The edge relies on the specific structural features of the ATOM market (Swing Pressure, FVGs) captured by the logit coefficients.

Components

• Recursive Median Filter (regime) — Provides a noise-filtered trend regime by ignoring price outliers that often trigger false SMC signals.
• TradeEase (direction) — Identifies institutional market structure (BOS/CHoCH) to ensure entries align with directional shifts.
• ATOM Coefficient Indicator (logit) (entry) — Provides a tactical momentum probability score specifically optimized for the ATOM 15m timeframe.
• Ehlers Detrended Synthetic Price (DSP) (exit) — Identifies the exhaustion of the local cycle component to time exits before trend reversals.
• Spread Display and Alert (risk) — Filters out trades during high-volatility/low-liquidity events where transaction costs exceed potential edge.
• silent-s1-box-new-width-151-v20-v1 (confirmation) — Used as a technical confirmation of the charting engine's box rendering stability before executing orders.
• Boxline (Consolidation Zones) (volatility_filter) — Filters for expansion phases by ensuring the entry occurs immediately following a breakout from a consolidation box.

Known failure conditions

• The ATOM Logit model P-value stays between 0.45 and 0.55 for extended periods during high volatility.
• TradeEase BOS/CHoCH signals frequently repaint on the 15m timeframe, leading to 'phantom' entries in backtests.
• The recursive median filter alpha calculation hits a division-by-zero state.

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.