LAB #1041 PROMISING: LEDGER BENCH: Apply filter: Min confirmations >= 5 — improves win rate from 31.5% to 33.1% (+1.6pp, — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 5 — improves win rate from 31.5% to 33.1% (+1.6pp, n=1326) SAME-WINDOW EFFECT: Applying min confirmations >= 5 to the live trade window would have filtered out several low-confidence entries. The REJECTED trades (XAUUSD SHORT, EURJPY SHORT, AUDUSD SHORT) already failed to open, so no change there. Among closed trades, NZDUSD LONG (currently opening) and GBPUSD LONG (loss) would likely be excluded if they had fewer than 5 confirmations, removing a loss and an unknown. The net effect on win rate is marginal here — roughly +0.5pp to +1.0pp improvement, as the sample is too small to replicate the backtest gain precisely. OVER-TIME PROJECTION: Over 3-6 months, this filter should reduce noise entries that trigger on weak signals, raising win rate by 1-2pp as seen in the backtest. However, it will also reduce total trade count by roughly 15-20%, potentially lowering absolute PnL if the Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com