Shapes before the 8am UTC funding print are not an edge
BTC 24h shapes before 08:00 UTC funding: holdout cluster C0 +3.3 bp, t=0.36 vs other clusters. NO_EDGE. Distinct from the already-killed funding-rate level.
The verified falsification record — reproducible trading strategies, out-of-sample stress tests, and negative findings published with complete code, seeds, and evidence.
BTC 24h shapes before 08:00 UTC funding: holdout cluster C0 +3.3 bp, t=0.36 vs other clusters. NO_EDGE. Distinct from the already-killed funding-rate level.
10-name crypto XSMOM R=20/H=5/K=2: val LS t=0.23, hit 46.5%; K=3 no rescue. NO_EDGE.
Crowded-long ∩ 15m vol filter: val n=38 mean +66 bp but t=1.41 — louder, not stronger than parent t=1.60. WATCH.
Feature Lab marks lsratio/HIGH actually_tradable. Neural: no incremental vs price MLP. Overlapping val t=4.10; nonoverlap t=1.60; WFE=0.16. WATCH, not a book.
BTC top-trader L/S HIGH: val fade t=−2.17 but corr vs global 0.99; incremental γ t=−1.45. Same costume. NO_EDGE.
BTC 15m volume residual |fwd r|: val and OOS both clear after |r| control (~+4–6 bps); discovery residual null. WATCH, not a trade.
BTC OI Δ: long-crowding fade fails val; extreme |ΔOI| lifts |fwd r| +31% (t=4.41), largely shared with prior |r|. WATCH.
BTC 5m volume residual: val γ t=+8.45, OOS +7.58; discovery also clears. WATCH magnitude leftover, not a trade.
BTC HIGH-only nonoverlap: n=71 val t=1.60 fails |t|≥2; mean still ~44 bp; OOS sign holds. Demote overlapping PROMOTED to WATCH.
BTC HIGH-only crowded-long short: val +41.5 bp at 5 bps (t=3.93), OOS sign holds; not redundant with OI. Clustered overlap caveat. PROMOTED research object — not live.
Pro DQN leftover+price-only both cancelled at 1800s (~ep 164/200), no model. Dummy overlapping val t=4.10; nonoverlap t=1.42. NO_EDGE.
Classic BTC pin-bar: continuation wrong-signed; fade-tilt is loc/wick costume (Q2). Dies at 5 bps. NO_EDGE.
ETH/SOL global L/S HIGH: val t=−2.06 / −3.13, not redundant with |r|; ETH OOS holds; Q3 ETH ≤0 at 0 bps. WATCH, not a book.
ETH 15m volume residual C vs A: val t_γ=+2.12, OOS +2.56, disc also clears (unlike BTC). ~4.7 bps extra |r|. WATCH, not a trade.
BTC funding→next 8h mark: val t(H−M)=−0.62; OOS n_high=0; mark vs spot corr 0.9999. Same null as spot-aligned funding. NO_EDGE.
BTC+ETH+SOL nonoverlap HIGH shorts: val n=730 mean +11.8 bp t=1.22 — weaker than BTC-only. ETH regime dilutes. WATCH.
ETH/BTC |z|≥2 residual MR: discovery continuation (t=+2.18); val non-overlap fade LS t=+1.55 n=9. NO_EDGE.
BTC |r| clustering is real (1h AC1 ~0.27) but half-life is unstable (29h vs 5h). Sign AC1 ~−0.07, not an edge. WATCH.