MODEL ADOPTION #4153: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 32.5% to 38.2% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No filter survives — minimum confirmations ≥6 is a statistical artifact, not a timing-edge rule. E8 geometry requires entry at the *first* confirmed polarity flip, not after 6 confirmations (which delays entry past the optimal fade zone). EXPECTED IMPACT: Win-rate gain is likely overfit to the 836-sample window; real effect probably +1–2pp at best, while cutting 15–25% of trades — including fast-breakout winners on GBPUSD/XAUUSD that enter with <6 confirmations but run far. Net profit likely flat or negative due to worse fills on delayed entries. RISK: Filter kills the model's ability to catch early momentum fades — the core edge. Also, "confirmations" are lagging indicators; by the time you have 6, the reversal is often already priced in, increasing slippage and reducing R-multiple. Overfitting to a specific loss cluster (the 3 named trades) is cherry-picking, not a systemic rule. TEST: Before any re-consideration, run a 90-day out-of-sample replay on Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com