MODEL ADOPTION #2862: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.1% to 37.6% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — the ≥6 confirmations filter is a statistical artifact, not a geometric timing edge. E8-node/Gods-Clock rhythm logic is phase-based; confirmation count is a lagging indicator that fights the fade polarity (you fade early, not late). EXPECTED IMPACT: Win-rate gain (+6.5pp) is real in backtest but likely from regime overfit — it will cut high-RR breakout fades that enter before confirmation, reducing net expectancy even if win% rises. Profit impact: negative or flat over 3–6 months. RISK: Filter rejects early entries in fast moves (the exact E8 fade timing), causing missed winners; also overfits to recent correlation/volatility regime — degrades when rhythm shifts. TEST: Must pass a walk-forward on rizzy5m only: split last 12 months into 6-month train/6-month test. Optimize threshold (4–8) on train, require threshold=6 to beat threshold=1 (no filter) on test by ≥3pp win rate AND ≥0.1R average R-multiple improvement. Also require that Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com