ATOM Logit Confluence Funnel

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

Thesis

The hypothesis posits that a machine-learning model with low individual predictive power (AUC 0.52) can be rendered profitable by applying it only when multi-layered technical confluence (IAE) and confirmed market structure (ZigZag) align, effectively filtering for 'high-conviction' momentum bursts in ATOMUSDT.

Components

• ZigZag (regime) — Used to establish the macro-trend context by identifying the last confirmed swing high/low sequence. It filters for entries that align with established market structure.
• ATOM Coefficient Indicator (logit) (direction) — Provides the primary directional bias using a specialized logistic regression model specifically tuned for ATOMUSDT 15m micro-structures.
• MACD Demo Implementation (entry) — Serves as the tactical entry trigger to ensure trade execution coincides with short-term momentum shifts.
• SuperTrend (exit) — Acts as a trailing stop-loss and trend-exhaustion exit mechanism.
• ATR Heiken Ashi (risk) — Determines the initial stop-loss distance and position size based on smoothed volatility, accounting for the 'noise' reduction of HA candles.
• IAE — Technical Confluence Score (confirmation) — Filters out low-probability signals from the Logit model by requiring a high composite score (confluence) across 10 technical layers.

Known failure conditions

• ATOMUSDT correlation with BTC exceeds 0.95 for extended periods, neutralizing ticker-specific model edges.
• The IAE Confluence Score remains in the 40-60 range for >100 bars, indicating a regime of indecision where the MACD trigger will produce excessive whipsaws.
• Realized slippage on ATOMUSDT exceeds the 0.05% FVG threshold used in the Logit features.

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).