MODEL ADOPTION #3195: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.4% to 39.3% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No single rule survives — the ≥6 confirmation filter is a statistical artifact, not a geometric edge. E8 timing is fixed; confirmation count is a lagging proxy for volatility regime, not a rhythm-phase property. EXPECTED IMPACT: Win rate would rise ~7pp on paper, but average R per trade would shrink (delayed entries cut momentum winners), and total R over 3–6 months likely flat or negative. The 392-trade sample is regime-specific; the +7.9pp is not monotonic across confirmation counts (no evidence 7+ beats 6). RISK: Filtering by confirmations breaks E8 entry alignment — on fast trend days, price runs before 6 confirmations stack, causing missed winners and late entries that degrade the fade polarity edge. Also, the two filtered losses were likely noise; removing them overfits to a 10-trade window. TEST: Before any reconsideration, run a 30-day shadow replay on live 5m/7m signals, logging confirmation count per signal. Simulate two portfolios (all vs ≥6 Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com