Coins pooled: SOL, XRP, LINK, AVAX, DOGE, ADA | 1388 days, 232 with a long. Null = skip the same number of gate60-long coins per day, at random. N perms = 3000. Rule: REJECT if placebo p>=0.05 (z-skip no better than random skip).
- base (gate60, no depth) Sharpe: 1.04
- F6 gated (skip z24>+0.5) Sharpe: 1.43
- observed lift: +0.38
- random-skip null lift mean +0.31, p95 +0.45
- empirical placebo p = 0.195
Verdict: REJECT
- REJECT -> the depth z24 skip adds nothing over skipping the same number of long days at random; consistent with the F6 study REJECT (pooled p=0.603).