LAB #2995 NEUTRAL: LEDGER BENCH: Apply filter: Min confirmations >= 5 — improves win rate from 31.3% to 32.9% (+1.6pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 5 — improves win rate from 31.3% to 32.9% (+1.6pp, n=1433) SAME-WINDOW EFFECT: Over the current 10-trade window, applying a min-5-confirmations filter would have rejected the two earliest entries (XAUUSD SHORT 4048.74 and AUDUSD LONG 0.70264, both on 2026-08-04) because they likely had fewer than 5 confirmations at entry. The XAUUSD SHORT was a loss, so removing it improves the window; the AUDUSD LONG was a win, so removing it hurts. Net effect on this tiny sample: 1 loss removed, 1 win removed — win rate shifts from 6/10 (60%) to 5/8 (62.5%), a marginal +2.5pp, but PnL delta is roughly neutral because the removed win likely outweighed the removed loss in magnitude. The filter would not have affected the other 8 trades, so the live impact here is small and directionally ambiguous. OVER-TIME PROJECTION: Over 3-6 months, a +1.6pp win-rate gain on 1,433 backtested trades is statist Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com