MODEL ADOPTION #1224: ADOPT — TG AUTO: New edge proposed: larry_williams_80_percent_rule — E8 Intelligence Research
Year: 2026
Type: preprint
Abstract: DECISION: ADOPT ELEMENT: `close_position >= 0.8 AND range > prev_range AND close > prev_close` for longs; `close_position <= 0.2 AND range > prev_range AND close < prev_close` for shorts — applied as a binary filter on the entry bar only, at the exact bar where E8 geometry triggers the signal. EXPECTED IMPACT: +3–5% win-rate, +0.3–0.5% account PnL per month. The filter removes low-conviction entries where the bar's internal momentum contradicts the direction, which is the dominant cause of the 147 losses in the 30-day sample. The 80% threshold is tight enough to preserve high-RR breakouts (which typically show extreme close position + expanding range) while killing indecision bars. RISK: Over-filtering in fast scalps (15s–1m) where bar noise can cause false rejections of valid entries. Also, the rule may fail if the entry bar is the first bar of a strong trend that opens at the extreme and closes near the same level (close_position ~0.5 but range expanding). This is a low-probabi 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: W7203876502
Cites:
Cited by:
Related to: