MODEL ADOPTION #2307: REJECT — LEDGER BENCH: Apply filter: XAUUSD SHORT — improves win rate from 31.5% to 50.0% (+18.5pp, — E8 Intelligence Research

DECISION: REJECT ELEMENT: No filter survives — the XAUUSD SHORT-only rule is a regime artifact, not an E8-geometry edge. EXPECTED IMPACT: Win-rate gain (+18.5pp) is real but fragile; over 3–6 months, if gold trends up, blocking longs removes profitable trades, net PnL likely negative. The 32-trade sample is too small to separate signal from noise, and the filter does not align with E8-node/Gods-Clock rhythm logic — it's a price-level heuristic, not a timing edge. RISK: Overfitting to a recent bearish gold phase; filter becomes a drag in a rally. Also, it ignores systemic losses in other pairs (e.g., GBPUSD SHORT), so it doesn't fix the core model weakness. TEST: Before any reconsideration, run a 12-month full-history backtest (not just 32 trades) requiring ≥50 XAUUSD signals per regime (bull/bear/range). Pass only if: (a) win-rate improvement >10pp in both bear and range regimes, (b) no win-rate degradation in bull regime >5pp, (c) max drawdown reduced by ≥20% vs unfiltered. If Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com