MODEL ADOPTION #3767: REJECT — LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 32.3% to 38.6% — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — the 6-confirmation filter is a statistical artifact, not a geometric edge. EXPECTED IMPACT: Win-rate gain (+6.3pp) is real in backtest but likely from removing early entries that already had poor R:R in that sample; over live, it will cut both losers and high-momentum winners (e.g., XAUUSD news spikes), netting near-zero PnL change while increasing average holding time and stop distance. RISK: Filter delays entries by >2 bars in fast regimes, causing missed fills or worse prices; also, 6 is not anchored to E8 rhythm (Gods-Clock harmonics), so it's a data-mined threshold that will decay under regime shift. TEST: Before any reconsideration, run a 90-day shadow replay comparing filtered vs unfiltered on: (1) average R-multiple per trade, (2) profit factor, (3) max drawdown, (4) entries delayed >2 bars — require filtered profit factor ≥1.15 AND unfiltered profit factor ≤1.05, plus no more than 10% of filtered trades delayed >2 bars. I Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com