MODEL ADOPTION #3871: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 5 — improves win rate from 32.5% to 34.0% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No single rule survives — the 5-confirmation filter is a lagging heuristic that conflicts with E8's rhythm-based entry timing (Gods-Clock polarity fires at specific fractal nodes; adding confirmations delays execution past the optimal fade point). EXPECTED IMPACT: Win-rate gain (+1.5pp backtested) is real but likely regime-dependent; in live E8, delayed entries will convert fast breakouts into late fades, reducing R:R and increasing average loss per trade. Net PnL impact is near zero or negative after accounting for 10–15% fewer trades and worse entry prices. RISK: Overfits to choppy regimes; in trending markets, the filter systematically misses the first 5 candles of momentum, turning would-be winners into break-even or small losses. Also, the +7.5pp "same-window" effect is an artifact of removing two losing trades — not a stable edge. TEST: Replay-bench on last 6 months of E8 live data (all rizzy5m/rizzy7m/wkjudas signals) with filter thresholds 4, 5 Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com