LAB #2786 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 30.9% to 37.2% (+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 30.9% to 37.2% (+6.3pp, n=422) SAME-WINDOW EFFECT: Applying a minimum 6-confirmation filter would have eliminated the two XAUUSD SHORT losses (entries at 4048.74 and 4051.79) and the GBPUSD SHORT loss (entry 1.34508) — all three show entry timestamps within 15 minutes of a prior signal, suggesting low confirmation. That removes 3 losses from 10 trades, flipping the live window from 6W/4L to 6W/1L (if the rejected USDCHF stays open), a win-rate jump from 60% to ~86% on closed trades. PnL delta is strongly positive, roughly +2.5R to +3R, assuming average loss size. However, the filter would also have blocked the winning AUDUSD LONG (rizzy7m, 11:23) if applied across all timeframes — so the effect is sensitive to whether the filter is timeframe-agnostic. OVER-TIME PROJECTION: Over 3–6 months, the +6.3pp win-rate gain in backtests likely holds if the edge Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com