LAB #1860 NEUTRAL: LEDGER BENCH: Apply filter: EURUSD LONG — improves win rate from 32.0% to 35.6% (+3.6pp, n=45) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: EURUSD LONG — improves win rate from 32.0% to 35.6% (+3.6pp, n=45) SAME-WINDOW EFFECT: Over the current live window, applying a EURUSD LONG-only filter would have blocked the single EURUSD SHORT (rizzy5m, entry 1.13594, still open) — that trade is currently neither a win nor a loss, so removing it would not change the win rate (still 4/7 = 57.1%) but would reduce exposure and potential downside. The filter would not affect any other trades since none are EURUSD LONG. Expected PnL delta is near zero in this window, but it would eliminate one open risk position, slightly reducing variance. The claimed +3.6pp improvement is not visible here due to the tiny sample (n=45 backtest vs. 1 live EURUSD trade). OVER-TIME PROJECTION: Over 3–6 months, a EURUSD LONG-only filter could plausibly improve win rate if the backtest signal is real, but the effect size (+3.6pp on n=45) is within noise — 95% confidence in Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com