LAB #3839 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 32.5% to 38.3% (+5.8pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 6 — improves win rate from 32.5% to 38.3% (+5.8pp, n=807) SAME-WINDOW EFFECT: Applying the min-6-confirmations filter to the 10 live trades above would have rejected the 3 rizzy5m entries (AUDUSD, NZDUSD, XAUUSD) that all occurred within 2 hours of each other on 2026-08-27 — two of which closed as losses and one remains open. This would have removed 2 confirmed losses and 1 open position, leaving 7 trades with 2 wins and 2 losses (plus 3 open/unknown). The immediate win-rate would shift from 2/7 (28.6%) to 2/4 (50%) on closed trades, but the sample is too small to attribute this to the filter — it's likely noise from a single bad day. OVER-TIME PROJECTION: The +5.8pp win-rate improvement from n=807 backtests suggests a modest but real edge, likely by filtering out low-conviction entries that occur before sufficient confirmation stacking. Over 3-6 months, this could translate to Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com