← all research
  • Universe: BTC, ETH, SOL, XRP, LINK, AVAX, DOGE, ADA
  • Split (fixed): IS < 2025-07-01, OOS >= 2025-07-01
  • Costs: 8 bps/side, lag-1, long/flat only
MechanismSegRetSharpeMaxDDPFdaysboot p
N1 Dual-momentum top2/126dIS+28.5%1.04-7.5%1.21943nan
N1 Dual-momentum top2/126dOOS-0.6%-0.06-7.2%0.994450.524
N2 Low-vol top2/30dIS+12.6%0.74-6.4%1.12943nan
N2 Low-vol top2/30dOOS-1.1%-0.10-10.1%0.994450.554
N3 Turn-of-month EW8IS+0.9%0.12-5.5%1.05943nan
N3 Turn-of-month EW8OOS+1.1%0.30-2.8%1.114450.368
N4 52w-high top2IS+11.3%0.64-7.6%1.12943nan
N4 52w-high top2OOS-0.9%-0.08-9.7%0.994450.540

Net Sharpe by calendar year

Mechanism20222023202420252026full
N1 Dual-momentum top2/126d0.001.511.40-0.111.030.79
N2 Low-vol top2/30d0.001.460.590.15-0.320.46
N3 Turn-of-month EW81.091.45-0.670.040.830.17
N4 52w-high top20.001.250.92-0.330.010.42

Verdict bar: pooled OOS NET Sharpe > 0, and positive in the bear years (2025-26) to prove it is not just long-crypto beta.

Verdict: 0/4

  • N1 Dual-momentum — the absolute-momentum gate does its job on risk (OOS maxDD -7.2% vs -38% for plain TS momentum) but OOS return is -0.6% (SR -0.06, boot p 0.524): capital protection, not an edge.
  • N2 Low-vol OOS SR -0.10; N4 52w-high OOS SR -0.08. Both dead.
  • N3 Turn-of-month is the only positive OOS path (+1.1%, SR 0.30) but it does not survive scrutiny: OOS bootstrap p = 0.368, and a matched-exposure control (3000 random days at the same 17% duty cycle) gives p = 0.340 — i.e. indistinguishable from randomly sitting in cash. 2024 was negative.

Combined with batch 1 (6/6 reject), 10 pre-registered long-only crypto mechanisms now fail the held-out test. Alpaca paper crypto is long-only, so no market-neutral (short-leg) construction is tradable on this account — which is the structural reason a long/flat crypto bot cannot clear a bear.