LAB #4509 NEUTRAL: LEDGER BENCH: Apply filter: XAUUSD LONG — improves win rate from 31.6% to 54.1% (+22.5pp, n=85) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: XAUUSD LONG — improves win rate from 31.6% to 54.1% (+22.5pp, n=85) SAME-WINDOW EFFECT: Applying an XAUUSD LONG-only filter would have zero effect on the current trade window, as none of the 10 listed live trades involve XAUUSD. The filter does not touch the USDCHF, EURJPY, USDCAD, AUDUSD, GBPUSD, NZDUSD, or USDJPY positions shown. Therefore, the claimed +22.5pp win-rate improvement is entirely out-of-sample relative to this window — it cannot be validated or refuted by these trades. Expected PnL delta here is exactly zero. OVER-TIME PROJECTION: Over 3-6 months, a LONG-only XAUUSD filter could meaningfully reduce drawdown if the underlying edge is real, but the sample (n=85) is small and likely cherry-picked from a specific regime. Gold has strong directional persistence, but also sharp mean-reversion spikes; a LONG-only filter will miss profitable SHORT opportunities and may overfit to a bullish pe Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com