LAB #3964 NEUTRAL: LEDGER BENCH: Apply filter: Min confirmations >= 5 — improves win rate from 32.5% to 33.9% (+1.5pp, — E8 Intelligence Research

Abstract

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 5 — improves win rate from 32.5% to 33.9% (+1.5pp, n=1812) SAME-WINDOW EFFECT: Over the current live window, applying a minimum 5-confirmation filter would have removed the two rizzy5m entries (AUDUSD SHORT, NZDUSD SHORT) and the rizzy5m XAUUSD SHORT, all of which had fewer than 5 confirmations at entry. The AUDUSD is still open (rejected), the NZDUSD closed as a loss, and XAUUSD closed as a loss — so filtering them out would have removed 2 losses and 1 open/rejected trade. The remaining 7 trades would have kept 2 wins and 3 losses, shifting the live win rate from 2/7 (28.6%) to 2/4 (50%) on closed trades, but the small sample makes this delta statistically meaningless. PnL impact is likely positive but trivial in magnitude given the tiny n. OVER-TIME PROJECTION: Over 3–6 months, the +1.5pp win-rate improvement in backtests (n=1812) is plausible but marginal — it likely comes fr Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com