LAB #4607 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 5 — improves win rate from 31.6% to 33.1% (+1.4pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 5 — improves win rate from 31.6% to 33.1% (+1.4pp, n=1902) SAME-WINDOW EFFECT: Applying a minimum of 5 confirmations would have filtered out the three REJECTED/open trades (USDCHF, EURJPY, AUDUSD) — all of which are currently unresolved and would have been excluded from the live book. Among the seven closed trades, none would have been removed, so the closed-trade win rate stays at 4/7 (57.1%). The net effect is a cleaner book with fewer open exposures, but no measurable PnL delta on closed trades; the +1.4pp backtest gain is not visible in this tiny window because the filter only removes pending/rejected signals, not losing closed ones. OVER-TIME PROJECTION: Over 3–6 months, the filter likely reduces trade frequency by roughly 15–20% (based on the REJECTED ratio here) while improving average signal quality. The risk is that it may also cut early winners that would have closed i Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com