LAB #4178 NEUTRAL: LEDGER BENCH: Apply filter: Min confirmations >= 5 — improves win rate from 32.6% to 34.1% (+1.5pp, — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 5 — improves win rate from 32.6% to 34.1% (+1.5pp, n=1825) SAME-WINDOW EFFECT: Over the current live window, applying a min-5-confirmations filter would have rejected the three rizzy5m entries that fired within 5 minutes of bar close (GBPUSD 11:50, NZDUSD 10:20, AUDUSD 10:15 on 2026-09-02) — all three were entered at 5-minute candle closes with confirmations likely below 5. Of those, GBPUSD and NZDUSD were wins, AUDUSD was a loss — so the filter would have removed 2 wins and 1 loss, reducing win count from 3 to 1 in that subset. Net PnL delta is negative for this window (−2 wins), but the sample is tiny and dominated by wkjudas losses that the filter does not touch. The backtest gain of +1.5pp is not visible here; it would have hurt this specific window. OVER-TIME PROJECTION: Over 3–6 months, the +1.5pp win-rate gain (n=1825) suggests the filter removes low-quality early entries Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com
Author: Caldin, Andrew Stewart
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7222868933
Cites:
Cited by:
Related to: