LAB #2978 PROMISING: LEDGER BENCH: Apply filter: XAUUSD LONG — improves win rate from 31.3% to 58.1% (+26.8pp, n=74) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: XAUUSD LONG — improves win rate from 31.3% to 58.1% (+26.8pp, n=74) SAME-WINDOW EFFECT: Applying a strict XAUUSD LONG-only filter would have blocked the three XAUUSD SHORT trades in the window (all losses: 4048.74, 4051.79, and the 4270.17 LONG would remain). That removes 3 losing trades from the 10 listed, leaving 7 trades with 5 wins and 2 losses — a live win rate of ~71.4% vs. the unfiltered 50% (5/10). PnL delta is positive but modest: you avoid three full losses, but you also forfeit any future short-side winners that might appear in the same window. The single XAUUSD LONG that was already a win stays, so the immediate effect is a clean reduction of drawdown, not a massive profit boost. OVER-TIME PROJECTION: Over 3–6 months, this filter will likely raise overall win rate and reduce variance, but it may also cut total trade count and cap upside if gold trends bearish for extended periods. The ba Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com