LAB #2568 PROMISING: LEDGER BENCH: Apply filter: XAUUSD SHORT @ NY_OPEN with 5 conf — improves win rate from 31.0% to 58. — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: XAUUSD SHORT @ NY_OPEN with 5 conf — improves win rate from 31.0% to 58.3% (+27.3pp, n=12) SAME-WINDOW EFFECT: Over the current 10-trade window, the filter would have blocked the XAUUSD SHORT entries on 2026-08-04 and 2026-07-31 (both LOSSES), leaving the XAUUSD LONG untouched. That removes 2 losing trades from the sample, shifting the window from 6W/4L (60% win) to 6W/2L (75% win) — a +15pp improvement purely from filtering out the two worst XAUUSD shorts. However, the filter would also have blocked any potential XAUUSD SHORT that might have been taken at NY_OPEN on other days — none appear here, so no missed winners in this window. Net PnL delta is positive: +2 losses avoided, no wins sacrificed. OVER-TIME PROJECTION: Over 3–6 months, the filter's value depends on whether the 58.3% win rate (n=12) is stable or a small-sample artifact. If XAUUSD SHORT at NY_OPEN with 5 conf is genuinely a low-quali Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com