LAB #4960 NEUTRAL: LEDGER BENCH: Apply filter: Min confirmations >= 7 — improves win rate from 31.7% to 48.1% (+16.4pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 7 — improves win rate from 31.7% to 48.1% (+16.4pp, n=27) SAME-WINDOW EFFECT: The filter would have blocked the NZDUSD, GBPUSD, EURUSD(rizzy7m), USDCHF, and EURJPY entries — all currently REJECTED/open, so no closed-trade PnL changes. Of the 5 closed trades in the window (3 losses, 2 wins), the filter's effect depends on their confirmation counts, which aren't given; if the 3 losers had <7 confirmations they'd be removed, lifting the window from 40% to ~100% on a 2-trade sample. That's the tell: the claimed +16.4pp is on n=27, and the live window is far too small to confirm or refute it. OVER-TIME PROJECTION: If the +16.4pp holds, expectancy improves materially and the filter is worth keeping — but n=27 with a 7-confirmation threshold is a classic overfit candidate: you're selecting on a feature after seeing outcomes, and the surviving sample is tiny. Over 3-6 months, expect reg Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com