LAB #2044 NEUTRAL: LEDGER BENCH: Apply filter: USDCHF SHORT — improves win rate from 32.0% to 59.8% (+27.8pp, n=169) — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: USDCHF SHORT — improves win rate from 32.0% to 59.8% (+27.8pp, n=169) SAME-WINDOW EFFECT: Over the current live window, the USDCHF SHORT filter would have had **zero direct impact** — none of the 10 listed trades involve USDCHF. The filter is symbol-direction-specific, so it would not alter any of these outcomes. However, if applied as a global exclusion rule (blocking all USDCHF SHORTs), it would only matter if such a trade had been taken; none were. Expected PnL delta in this window: **0%**. The win-rate improvement cited (32%→59.8%) is from a backtest subset (n=169) that does not overlap with this live sample, so no immediate effect is observable here. OVER-TIME PROJECTION: Over 3–6 months, this filter could meaningfully reduce losses if the USDCHF SHORT edge is real and stable. However, the sample size (n=169) is small, and the +27.8pp jump is suspiciously large — likely overfit to a specific re Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com