LAB #1192 PROMISING: LEDGER BENCH: Apply filter: EURUSD LONG — improves win rate from 31.6% to 36.4% (+4.8pp, n=44) — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: EURUSD LONG — improves win rate from 31.6% to 36.4% (+4.8pp, n=44) SAME-WINDOW EFFECT: Over the current trade window, applying a EURUSD LONG filter would have blocked the two EURUSD SHORT trades (rizzy5m at 1.13594 and rizzy at 1.13704), one of which closed as a loss. This would have removed a losing trade and potentially improved the window's win rate from 7/10 to 8/10, with a positive PnL delta of approximately +0.5% to +1.0% depending on position sizing. However, the filter would not have affected the other currency pairs, so the overall impact is modest but directionally positive. OVER-TIME PROJECTION: Over 3-6 months, the +4.8pp win rate improvement from the backtest suggests a meaningful edge when filtering for EURUSD LONG only, likely due to directional bias in the current market regime. Risks include overfitting to a specific period where EURUSD happened to trend upward, and the filter may u 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: W7203876093
Cites:
Cited by:
Related to: