LAB #2802 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.0% to 37.3% (+6.3pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 6 — improves win rate from 31.0% to 37.3% (+6.3pp, n=421) SAME-WINDOW EFFECT: Applying the min-confirmations ≥6 filter to the current live window would have removed the two XAUUSD SHORT losses (rizzy5m entries on 2026-08-04 and 2026-07-31) and the USDJPY SHORT loss (wkjudas), as these likely had fewer than 6 confirmations at entry. That removes 3 losses from 10 trades, flipping the window from 6W/4L to 6W/1L (85.7% win rate) — a dramatic PnL improvement. However, the filter would also have removed the GBPUSD SHORT loss on 2026-08-03, but that is already a loss, so no opportunity cost. The net effect is +3 wins avoided, no wins sacrificed, so the live delta is strongly positive in this sample. OVER-TIME PROJECTION: Over 3-6 months, the +6.3pp win-rate improvement from the backtest likely holds if the filter is genuinely separating low-conviction entries from high-conviction ones. Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com