LAB #2463 NEUTRAL: LEDGER BENCH: Apply filter: Min confirmations >= 7 — improves win rate from 31.1% to 36.4% (+5.3pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 7 — improves win rate from 31.1% to 36.4% (+5.3pp, n=11) SAME-WINDOW EFFECT: Applying the min-confirmations ≥7 filter would have excluded the two rizzy5m losses (GBPUSD 1.34508, XAUUSD 4051.79) and the rizzy7m win (AUDUSD 0.70264) — but also the rizzy5m win (USDCAD 1.40358). Net effect on the 10-trade window: 6 wins/4 losses becomes 5 wins/3 losses (win rate 60%→62.5%), but PnL delta is roughly flat because the excluded win (USDCAD) likely had a larger TP2 gain than the two excluded losses combined. The +5.3pp backtest improvement does not replicate here — sample is tiny and dominated by instrument-specific variance. OVER-TIME PROJECTION: Over 3–6 months, a higher confirmation threshold should reduce false entries from noise-driven rizzy5m signals, likely improving win rate modestly (2–4pp) but also cutting total trade count by ~20–30%, which may reduce absolute PnL if wins are Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com