MODEL ADOPTION #817: REJECT — LEDGER BENCH: Apply filter: Pool density HIGH (>=90) — improves win rate from 32.1% to 33. — E8 Intelligence Research

Abstract

DECISION: REJECT ELEMENT: The filter "Pool density HIGH (>=90)" is not a concrete, testable edge compatible with E8 geometry timing. It is a statistical artifact from a moderate backtest (n=1289) that conflates correlation with causation. The E8 rhythm logic and Gods-Clock timing are deterministic — adding a density filter that is not derived from the same geometric framework introduces a non-E8 variable that can decouple the model from its core timing edge. The "same-window effect" cited is a cherry-picked 10-trade sample, not a robust projection. The 1.6pp win-rate lift is within noise for a 1289-sample backtest, and the risk of rejecting high-volatility winners (which E8 often targets) is unquantified but material. EXPECTED IMPACT: Negligible to negative. Win-rate might lift ~1pp in favorable regimes, but trade count could drop 15-25%, reducing absolute PnL. The filter has no timing anchor — it will reject trades that E8 geometry would otherwise time correctly, especially during l Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com