LAB #4928 NEUTRAL: LEDGER BENCH: Apply filter: ETHUSD SHORT @ NY_OPEN with 6 conf — improves win rate from 31.7% 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 31.7% to 63.6% (+31.9pp, n=11) SAME-WINDOW EFFECT: The filter is ETHUSD-specific, and the current live window contains zero ETHUSD trades — it's all EURUSD, XAUUSD, AUDUSD, NZDUSD, GBPUSD, USDCHF, EURJPY, USDCAD. So over this window the filter would have changed nothing: same 2 wins, 3 losses, 4 rejected/open, identical PnL. The claimed +31.9pp is drawn from a different instrument and a different sample than anything the live book actually traded, so it cannot be validated or falsified against this window. OVER-TIME PROJECTION: If the effect is real, it's a narrow, conditional edge (one pair, one session, one confidence bucket) and would add a small number of high-quality ETHUSD shorts per month — meaningful only if ETHUSD is a material share of your book, which the live window suggests it isn't. The dominant risk is n=11: a 63.6% win rate Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com