LAB #4413 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.6% to 35.7% (+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 31.6% to 35.7% (+4.1pp, n=910) SAME-WINDOW EFFECT: Applying a minimum-6-confirmation filter over the current live window would have rejected the three trades marked [REJECTED/open] — EURJPY SHORT, AUDUSD LONG, and NZDUSD SHORT (wkjudas). Of those, only EURJPY SHORT and AUDUSD LONG were actually rejected by the filter; NZDUSD SHORT (wkjudas) is still open and would have been blocked, removing an unknown outcome. Among the remaining seven trades, the win/loss split is 4 wins, 3 losses — identical to the unfiltered set, so no immediate win-rate improvement is visible in this tiny sample. The +4.1pp backtest gain likely comes from filtering out low-confidence setups across a larger population, but here the effect is neutral-to-slightly-positive at best, with no PnL delta calculable from the closed trades alone. OVER-TIME PROJECTION: Over 3–6 months, a stri Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com