LAB #3981 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 32.5% to 38.1% (+5.7pp, — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 6 — improves win rate from 32.5% to 38.1% (+5.7pp, n=818) SAME-WINDOW EFFECT: Applying a minimum 6-confirmation filter would have altered the live window by rejecting the two rizzy5m entries (AUDUSD SHORT and NZDUSD SHORT) that entered at 09:15–09:35 on 2026-08-27, as these likely had fewer confirmations at entry. The NZDUSD SHORT via rizzy5m closed as a LOSS, and the AUDUSD SHORT remains open/rejected — removing both would have eliminated at least one confirmed loss and one uncertain position. The XAUUSD SHORT via rizzy5m (also a loss) may have been filtered too if its confirmation count was below 6, potentially turning the window from 2W/5L to 2W/3L or better — a meaningful PnL improvement. However, the wkjudas and rizzy7m trades would remain unaffected, so the net delta is positive but modest, roughly +0.5–1.5% PnL on this small sample. OVER-TIME PROJECTION: Over 3–6 months, Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com