H-001: Minute-level TA ensemble (RSI, MACD, EMA, momentum) predicts 1-4 h direction

Fail Family: intraday · Status: rejected

Do RSI, MACD and EMA predict crypto's next hours? The idea did not beat simply buying the coin and holding it. It is rejected.

Result

Replay of 7,649 live signals (Jun-Sep 2026) on real prices: win rate 33.9% at 2:1 R:R, EV -0.019%/trade before fees; direction worse than random in all 12 TP/SL x timeout scenarios (-1.7..-3.8 sigma). Meta-labelling AUC 0.53.

Sharpe: return per unit of risk (higher is better). CAGR: average yearly growth. Max drawdown: worst fall from a peak. p-value: chance the result is luck (lower is better). “Hold” columns are buy-and-hold for the same asset and days. Glossary

No engine results yet: this study is kept for the record and is not re-run.

Why it might work

Original F1MEBOT strategy: ensemble of classical indicators on 1-minute candles, TP 0.8% / SL 0.4%.

Verdict history

2026-09-30: Fail (F1MEBOT (2026))