MODEL ADOPTION #4473: REJECT — LEDGER BENCH: Apply filter: Pool density HIGH (>=90) — improves win rate from 31.5% to 33. — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — pool density ≥90 filter is not a concrete, testable edge compatible with E8 geometry timing; it is a lagging regime proxy with no causal link to FADE polarity or Gods-Clock phase. EXPECTED IMPACT: Win-rate gain (+1.4pp) is real in backtest but likely overfit to recent volatility clustering; frequency loss (20–30%) would cut absolute PnL, and the live window shows benefit is concentrated in wkjudas signals only — not generalizable across providers. Net expected impact: wash to slightly negative. RISK: Density is lagging — filters out fresh-trend entries (E8's early FADE setups); threshold drift with regime; reduces trade count, starving the model of compounding opportunities; masks signal-source quality differences. TEST: No replay-bench test can pass — the filter must be rejected outright because it does not align with E8 geometry timing (density is not a phase/rhythm variable). If you insist on testing, run 300 shadow trades with

Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com