LAB #3848 NEUTRAL: LEDGER BENCH: Apply filter: XAUUSD SHORT — improves win rate from 32.5% to 47.1% (+14.5pp, n=34) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: XAUUSD SHORT — improves win rate from 32.5% to 47.1% (+14.5pp, n=34) SAME-WINDOW EFFECT: Applying the XAUUSD SHORT filter would have blocked the single XAUUSD SHORT trade in the window (entry 4598.79, CLOSED/LOSS). That removes one loss from the 10-trade set, shifting the live win rate from 3/10 (30%) to 3/9 (33.3%) — a modest improvement, not the +14.5pp claimed in backtest. PnL delta is small but positive, roughly equal to the loss magnitude of that one trade (likely 1–2% risk). No other trades are affected, so the immediate impact is narrow and directionally favorable but statistically trivial. OVER-TIME PROJECTION: Over 3–6 months, the filter would only apply to XAUUSD SHORT signals — a subset of total trades. If the backtest edge is real (n=34 is small, so likely inflated by variance), you might see a few percentage points of win-rate improvement, but the sample size is too small to separate si Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com