LAB #4980 NEUTRAL: LEDGER BENCH: Apply filter: Quarterly phase = ACCUMULATION — improves win rate from 31.7% to 44.2% ( — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Quarterly phase = ACCUMULATION — improves win rate from 31.7% to 44.2% (+12.5pp, n=138) SAME-WINDOW EFFECT: The filter would only bite on trades whose entry timestamp falls in a quarter classified ACCUMULATION — and the live window above spans 2026-09-03 to 2026-09-11, i.e. a single quarter-phase bucket. So the effect is binary, not selective: either the whole window is filtered out (all 8 trades suppressed, including the two USDCHF/USDCAD wins) or none of it is. Given the window's realized record is 2W/3L/3 rejected, a blanket suppression would have removed the two wins as well as the three losses — net PnL roughly flat-to-negative, not the +12.5pp the backtest advertises. The claimed improvement cannot be demonstrated on this window because there is no within-window contrast to exploit. OVER-TIME PROJECTION: If ACCUMULATION genuinely carries a 44.2% win rate vs 31.7% elsewhere, the filter is a reg 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: W7225007697