Scalping — CVD-Divergence Fade E5

Bybit perps, 60d 5m bars, UTC days · ATR≥0.5% universe

Entry-Filter Testing

Candidate entry filters tested against trade R: Spearman rank correlation + chronological 70/30 train/test split. "held up" = filter direction consistent out-of-sample.

MetricnρpSig?Train (filter vs other)Test (filter vs other)Held up?
CVD magnitude1092360.0010.7425n.s.+0.03R vs +0.04R+0.05R vs +0.08Rheld up
Direction (long=1)109236-0.0010.6641n.s.+0.04R vs +0.00R+0.07R vs +0.00Rheld up
Hour of day (UTC)1092360.0010.75n.s.+0.05R vs +0.02R+0.08R vs +0.05Rheld up
Read: with very large n, small ρ is statistically significant but practically negligible. Focus on filters that both reach significance AND show a meaningful train/test gap that holds direction.