LAB #3878 PROMISING: LEDGER BENCH: Apply filter: ETHUSD SHORT @ NY_OPEN with 6 conf — improves win rate from 32.5% to 63. — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: ETHUSD SHORT @ NY_OPEN with 6 conf — improves win rate from 32.5% to 63.6% (+31.2pp, n=11) SAME-WINDOW EFFECT: Over the current live window, the filter would have blocked the ETHUSD SHORT only if it had been present — but no ETHUSD trade appears in the live list, so direct impact is nil. However, applying the same logic (SHORT @ NY_OPEN with conf≥6) to the existing SHORTs would have filtered out the AUDUSD SHORT (rejected anyway), the NZDUSD SHORT (loss), and the XAUUSD SHORT (loss) — potentially avoiding 2 losses. The GBPUSD SHORTs (both wins) would have passed, so win rate on the filtered subset would rise from ~40% (4/10) to ~67% (2/3), but the sample is tiny and the PnL delta is ambiguous because the filtered-out losses were small vs. the wins. Net effect: modest improvement in win rate, negligible PnL change given the small n. OVER-TIME PROJECTION: Over 3–6 months, a 6-conf SHORT-only filter on Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com