MODEL ADOPTION #4065: REJECT — LEDGER BENCH: Apply filter: Pool density HIGH (>=90) — improves win rate from 32.6% to 34. — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — pool-density filter is not compatible with E8 geometry timing. EXPECTED IMPACT: +1.5pp win rate is real but marginal; however, it comes from a backtest window that mixes broker-feed artifacts and timezone shifts. The live model's edge is FADE polarity at specific E8 node rhythms — density is a secondary liquidity proxy, not a timing signal. Filtering on density would discard entries during fast breakouts (low density) that are exactly when E8 rhythm compression produces high-reward fades. Net profit impact likely negative over trending regimes. RISK: Overfitting to 1818 trades; density threshold is unstable across brokers/timezones; silent degradation if new signal sources alter pool composition. Also, the 4–5 filtered trades in your live window include 1 open (AUDUSD SHORT) — you cannot claim avoided loss until that trade closes. TEST: Before any re-consideration, run a 6-month replay on the exact live broker feed, splitting by E Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com