MODEL ADOPTION #3898: REJECT — LEDGER BENCH: Apply filter: Pool density HIGH (>=90) — improves win rate from 32.5% to 34. — E8 Intelligence Research

DECISION: REJECT ELEMENT: No rule survives — pool-density filter is a statistical artifact, not an E8-geometry-compatible edge. EXPECTED IMPACT: +1.5pp win rate is within noise for n=1808 (σ≈1.1pp), and the filter likely removes low-frequency, high-volatility setups (rizzy5m) that carry outsized R-multiples. Net PnL could be negative even if win rate rises. RISK: Regime-dependence — pool density is a liquidity proxy, not a timing signal. In fast markets (news, rollover), density drops precisely when E8 polarity edges are strongest. Filtering those kills the fade edge's best entries. Also reduces trade count, lowering gross profit. TEST: If you insist, run a walk-forward on 12 months in 3-month blocks — require ≥300 trades per block with ≥+2.0pp win-rate improvement AND positive R-multiple delta in at least 3 of 4 blocks. Also require the filter to improve the 10 live trades above (not just remove losses) — if it removes a winner, reject. Until then, no change. Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com