LAB #4488 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.5% to 35.6% (+4.1pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 6 — improves win rate from 31.5% to 35.6% (+4.1pp, n=913) SAME-WINDOW EFFECT: Applying a minimum 6-confirmation filter would have rejected the two REJECTED/open trades (EURJPY SHORT, AUDUSD LONG) and likely the two wkjudas LOSS entries (GBPUSD LONG, USDJPY SHORT) if they lacked confirmations at entry. Among the remaining six trades, you'd have kept four wins and two losses (AUDUSD SHORT loss, NZDUSD SHORT open), yielding a win rate of 4/6 = 66.7% versus the unfiltered 5/10 = 50% — but the sample is tiny and the two rejected trades were both open, so PnL impact is indeterminate. The +4.1pp backtest gain likely comes from cutting low-conviction losers, but here it would also cut one potential winner (EURJPY SHORT) if it later closed in profit. OVER-TIME PROJECTION: Over 3–6 months, the filter should reduce trade frequency by roughly 20–30% while lifting win rate toward the 35.6% b Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com