LAB #878 NEUTRAL: LEDGER BENCH: Apply filter: XAUUSD LONG — improves win rate from 31.5% to 61.2% (+29.6pp, n=67) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: XAUUSD LONG — improves win rate from 31.5% to 61.2% (+29.6pp, n=67) **SAME-WINDOW EFFECT:** Applying the XAUUSD LONG filter would have rejected the XAUUSD SHORT trade (currently REJECTED/open) and any other XAUUSD shorts in the window. Since the only XAUUSD trade was a short that did not close, the filter would have no direct PnL impact on closed trades. However, it would prevent future XAUUSD shorts from being taken, which could have avoided losses if those shorts had been active. The win-rate improvement in backtests (+29.6pp) suggests the filter is strongly directional, but the live window shows no XAUUSD longs to benefit from the filter. **OVER-TIME PROJECTION:** Over 3-6 months, this filter would likely increase win rate on XAUUSD trades by eliminating losing shorts, but it introduces a significant risk: it biases the system to only take XAUUSD longs, which may miss profitable short opportu Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com