MODEL ADOPTION #3015: REJECT — LEDGER BENCH: Apply filter: Pool density HIGH (>=90) — improves win rate from 31.3% to 32. — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — pool density HIGH (≥90) is a lagging aggregate that does not align with E8-node/Gods-Clock phase geometry; it filters by crowd thickness, not by timing polarity. EXPECTED IMPACT: Win-rate gain (+1.6pp) is within noise (p≈0.08–0.12), and the 2-loss removal in the live window is anecdotal (n=10). Over 3–6 months, trade frequency drops 15–25% without proportional loss trimming — net profit likely flat to slightly negative after costs/slippage. RISK: Density is high during news spikes and crowded breakouts — exactly when E8 fade polarity is most vulnerable. Filter would remove low-density contrarian setups that are geometrically valid, degrading signal diversity and increasing correlation to herd behavior. Also, backtest overfits to recent volatility regime; live regression to mean is probable. TEST: Before any reconsideration, run a 6-month walk-forward (train 4 months, test 2) requiring: (a) ≥300 filtered vs ≥300 unfiltered trades, Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com