MODEL ADOPTION #2060: REJECT — LEDGER BENCH: Apply filter: Pool density HIGH (>=90) — improves win rate from 32.0% to 33. — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — the pool-density ≥90 filter is a statistical artifact, not an E8-geometry-compatible edge. EXPECTED IMPACT: +1.6pp win-rate on 1,351 trades is within noise (σ≈1.3pp for n=1351 at p=0.32). The 10-trade window delta is ±0.8R, indistinguishable from zero. Filtering 10–20% of signals reduces sample size, increasing variance and risking regime overfit — no evidence it aligns with Gods-Clock rhythm or FADE polarity timing. RISK: Degrades model by (a) cutting counter-trend trades that thrive in low-density pools (where FADE polarity often wins), (b) creating cliff effects at threshold 90 (sensitivity untested), (c) masking true edge with selection bias from a single backtest window. TEST: Must pass a 5-year, walk-forward replay on E8-node timestamps: split into 12 rolling 6-month windows, require ≥300 filtered vs unfiltered trades per window, and demand the filter improves Sharpe by ≥0.15 in at least 9/12 windows with no window worse tha Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com