LAB #4961 NEUTRAL: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.7% to 36.0% (+4.2pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 6 — improves win rate from 31.7% to 36.0% (+4.2pp, n=940) SAME-WINDOW EFFECT: The filter would have blocked entries with fewer than 6 confirmations, but the live window shows no confirmation-count data attached to any of the 10 trades — so the claimed +4.2pp cannot be verified against this sample. Of the 4 closed trades, 2 were wins (USDCHF, USDCAD) and 2 were losses (EURUSD, XAUUSD); if either win sat at ≤5 confirmations, the filter would have removed a winner and the net effect could be negative. The 6 REJECTED/open trades are unaffected in outcome terms but would need re-scoring to know if they'd have been filtered. Net: no demonstrable same-window improvement, and a real risk of winner-removal. OVER-TIME PROJECTION: A +4.2pp win-rate lift on n=940 is statistically meaningful (roughly 40 extra wins), but win rate alone is not edge — if the filter also cuts trade count materia Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com