LAB #4946 NEUTRAL: LEDGER BENCH: Apply filter: NZDUSD SHORT — improves win rate from 31.7% to 34.5% (+2.8pp, n=139) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: NZDUSD SHORT — improves win rate from 31.7% to 34.5% (+2.8pp, n=139) SAME-WINDOW EFFECT: Over the current window, the filter would have blocked exactly one live trade: NZDUSD SHORT via rizzy5m at 0.58415 on 2026-09-09, which is currently sitting REJECTED/open. That trade's outcome is unresolved, so the filter's effect on this window is not a win-rate improvement — it's the removal of one open position with unknown PnL. The other nine trades (EURUSD, XAUUSD, AUDUSD shorts, USDCHF longs, EURJPY, USDCAD) are untouched, so realized PnL in this window is unchanged; the only delta is one fewer open NZDUSD short. OVER-TIME PROJECTION: A +2.8pp win-rate lift on n=139 is a real but modest edge — roughly 4 extra wins in that sample, which is within noise for a binomial of that size (standard error on a 31.7% rate at n=139 is ~3.9pp, so +2.8pp is under one sigma). Over 3-6 months, if the effect is genuine, exp Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com