LAB #2349 NEUTRAL: LEDGER BENCH: Apply filter: NZDUSD SHORT @ NY with 5 conf — improves win rate from 31.2% to 60.0% (+ — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: NZDUSD SHORT @ NY with 5 conf — improves win rate from 31.2% to 60.0% (+28.8pp, n=15) SAME-WINDOW EFFECT: Over the current live window, the filter would have blocked only NZDUSD trades — none of the 10 listed trades are NZDUSD, so zero trades would have been affected. The claimed +28.8pp win-rate improvement is based on n=15, which is statistically fragile (95% CI roughly ±25pp), meaning the true effect could be anywhere from +3pp to +54pp. PnL delta over this window would be $0 — the filter is irrelevant to the current book. The real question is whether the 15 NZDUSD SHORT NY trades with conf≥5 were cherry-picked or represent a stable edge. OVER-TIME PROJECTION: Over 3–6 months, if the filter is real, you'd expect NZDUSD SHORT NY win rate to stay above 50% (vs 31.2% baseline) — but with only ~15 trades/month, you'd need 6+ months to reach n=90 for statistical significance. Risk: the 60% win rate ma Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com