LAB #2619 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.0% to 37.6% (+6.6pp, — 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 37.6% (+6.6pp, n=412) SAME-WINDOW EFFECT: Applying a minimum-6-confirmation filter would have removed the two `rizzy5m` entries on GBPUSD (2026-08-07 11:35 and 2026-08-03 11:00) and the XAUUSD SHORT (2026-08-04 11:35) — all three were `rizzy5m` signals with fewer than 6 confirmations. Removing them would have eliminated 2 losses and 1 win, shifting the live window from 5W/4L (55.6%) to 4W/2L (66.7%) — a +11.1pp win-rate improvement, but with a net PnL delta of roughly −1R (since the removed win was larger than the removed losses). The filter would not have touched the `rizzy7m` or `wkjudas` trades, so those remain unchanged. OVER-TIME PROJECTION: Over 3–6 months, this filter likely reduces trade frequency by ~20–30% (mostly low-conviction `rizzy5m` entries) while improving average win rate by ~6pp, as backtested. The main risk is overfitting t Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com