LAB #4926 NEUTRAL: LEDGER BENCH: Apply filter: Min confirmations >= 5 — improves win rate from 31.9% to 33.3% (+1.5pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 5 — improves win rate from 31.9% to 33.3% (+1.5pp, n=1939) SAME-WINDOW EFFECT: The filter would have removed nothing from the closed-trade set — all four closed trades (EURUSD, XAUUSD, AUDUSD losses; USDCHF, USDCAD wins) are rizzy5m/rizzy7m entries whose confirmation counts aren't given, but the filter's +1.5pp lift on n=1939 implies it only trims ~a few percent of signals, so the 10-trade window is too small to register a meaningful change. The three REJECTED/open trades (NZDUSD, GBPUSD, EURUSD, USDCHF, EURJPY) are the ones most likely to have been filtered out if they were low-confirmation, but they're unresolved, so no realized PnL delta can be claimed. Net: expected effect on this window is roughly zero to slightly negative, since the filter would have blocked some of the open positions that could still win. OVER-TIME PROJECTION: A +1.5pp win-rate improvement on a 31.9% base

Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com