LAB #4111 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 5 — improves win rate from 32.6% to 34.1% (+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 32.6% to 34.1% (+1.5pp, n=1829) SAME-WINDOW EFFECT: Applying the min-confirmations ≥5 filter to the current live window would have rejected the AUDUSD SHORT (rizzy5m, 2026-08-27, entry 0.71765) which is currently marked REJECTED/open — likely a false signal that would have been filtered out, improving the sample. It would also have removed the NZDUSD SHORT (rizzy5m, 2026-08-27, entry 0.59416, CLOSED/LOSS) and the XAUUSD SHORT (rizzy5m, 2026-08-27, entry 4598.79, CLOSED/LOSS), both of which had fewer than 5 confirmations. Removing those two losses would flip the window from roughly 2W/3L (excluding open) to 2W/1L, a meaningful PnL improvement. However, the filter would not affect the wkjudas trades (different entry logic), so the net effect is positive but concentrated on rizzy5m entries. OVER-TIME PROJECTION: Over 3-6 months, a +1.5pp win-rate gain fro Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com