BTC daily time-series momentum does not clear validation

Verdict: NO_EDGE · Asset/TF: BTCUSDT 1d · Sample: chronological 60/20/20

Hypothesis: classic TSMOM — sign of the past R-day return predicts the next H-day return. Frozen primary R=20 / H=5 (non-overlapping for Q1). Distinct from the rejected alt/BTC cross-sectional momentum and from the 1h thin-fade family.

What happened

Q1 (val, non-overlap, n=101): E[+]=+0.46%, E[−]=+1.73%, Δ=−1.27% (t=−1.07), long-short=−0.43% (t=−0.70), hit=42.6%. Gate fails (|t|<2) and is wrong-way for TSMOM (shorts would have beaten longs).

Q2: val up-regime LS>0 weak; down-regime LS<0 — a bull-bias scent only; cannot promote after Q1 fail.

Secondaries (R=20/H=1, R=60/H=20): neither clears val |t|≥2.

Trading

Q3 skipped.

Final Verdict

NO_EDGE

Do not ship a 20/5 BTC TSMOM bot from this sample. Daily trend-following folklore does not survive a frozen chronological gate here — and the validation tilt goes the wrong way.