LAB #2059 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 5 — improves win rate from 31.9% to 33.6% (+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 31.9% to 33.6% (+1.6pp, n=1352) SAME-WINDOW EFFECT: Applying a minimum-confirmation filter of 5 would have blocked the three XAUUSD SHORT entries (rizzy5m) that all closed as losses — those were the weakest signals in the window. It would also have blocked the GBPUSD SHORT loss from 2026-08-03, leaving only the AUDUSD LONG, USDCAD SHORT, EURJPY LONG, and GBPUSD LONG wins intact. Net effect: 4 wins, 0 losses, 1 open (USDJPY) — win rate jumps to 100% on closed trades, and PnL improves by roughly +3.2R (avoiding 4 losses at ~0.8R each). This is a dramatic same-window improvement, but the sample is tiny and heavily skewed by XAUUSD's recent chop. OVER-TIME PROJECTION: Over 3–6 months, the filter should reduce false entries from low-conviction signals, likely lifting win rate by 1–2pp as backtested, but it will also cut total trade count by ~15–20% (since m Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com