LAB #2994 NEUTRAL: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.3% to 38.1% (+6.9pp, — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 6 — improves win rate from 31.3% to 38.1% (+6.9pp, n=438) SAME-WINDOW EFFECT: Applying a minimum-6-confirmation filter to the current live window would have removed the two most recent trades (GBPUSD SHORT at 11:35 and USDCHF SHORT at 10:30) because they were opened within minutes of signal generation. The GBPUSD trade was a WIN, so removing it would reduce the win count from 5 to 4 out of 8 remaining closed trades, dropping the window win rate from 62.5% to 50%. The USDCHF trade was rejected/open, so no PnL impact there. Net effect: the filter would have *hurt* this specific window's win rate and PnL, contradicting the backtest improvement — a red flag that the +6.9pp may be period-specific or driven by a different trade mix. OVER-TIME PROJECTION: Over 3–6 months, a higher confirmation threshold should reduce false entries and improve win rate if the backtest is robust, but it Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com
Author: Caldin, Andrew Stewart
Language: English
Subfield: Finance
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7215924163
Cites:
Cited by:
Related to: