LAB #4406 NEUTRAL: LEDGER BENCH: Apply filter: XAUUSD SHORT — improves win rate from 31.6% to 44.7% (+13.2pp, n=38) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: XAUUSD SHORT — improves win rate from 31.6% to 44.7% (+13.2pp, n=38) SAME-WINDOW EFFECT: Applying a blanket XAUUSD SHORT filter would have had **zero effect** on the current live window — none of the 10 listed trades are XAUUSD. The claimed +13.2pp win-rate improvement (n=38) is from a different instrument and time slice; it does not transfer to the FX pairs shown. PnL delta over this window: **0.0%**. The filter is irrelevant to the trades you are currently running, so it cannot explain or fix any live performance issue. OVER-TIME PROJECTION: Over 3–6 months, a single-pair, single-direction filter based on 38 trades is statistically fragile — the margin of error is roughly ±15pp at 95% confidence, meaning the true effect could be anywhere from −2pp to +28pp. XAUUSD SHORT behavior is regime-dependent (correlated with USD strength, risk-off flows, and gold's own volatility cycles); a filter that work Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com