LAB #2520 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.1% to 37.7% (+6.7pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 6 — improves win rate from 31.1% to 37.7% (+6.7pp, n=416) SAME-WINDOW EFFECT: Applying a minimum-6-confirmation filter would have removed the two lowest-conviction entries in the current window: the USDJPY SHORT (wkjudas, loss) and the XAUUSD SHORT on 2026-08-04 (rizzy5m, loss). That removes 2 losses from 10 trades, flipping the live record from 5W/5L to 5W/3L — a win-rate jump from 50% to 62.5% in this small sample. The PnL delta is positive but modest, since the removed losses are likely smaller than the removed wins (if any were removed — none were). The filter would not have touched any of the 6 winning trades, so the improvement is real but driven by only 2 data points. OVER-TIME PROJECTION: Over 3–6 months, a 6-confirmation minimum should raise win rate by roughly 5–7pp (consistent with your backtest), but it will also reduce trade frequency — likely by 20–30% — because ma Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com