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.
| Metric | n | ρ | p | Sig? | Train (filter vs other) | Test (filter vs other) | Held up? |
|---|---|---|---|---|---|---|---|
| Direction (long=1) | 200000 | 0.006 | 0.0074 | p<0.05 | +0.35R vs +0.35R | +0.37R vs +0.37R | did not hold |
| Hour of day (UTC) | 200000 | 0.066 | 0.0 | p<0.05 | +0.36R vs +0.34R | +0.38R vs +0.36R | held 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.