Pre-registered proposal: reports/research_proposal_carry_basis_overlay.md. Defaults: look=168h, carry|z|>=1.5, basis|z|>=2.0, multipliers {0.5, 1.0, 1.25}. Window: 2024-06-04 .. 2026-07-31 (788 days, 141 DOGE-gate60 long days) | Hyperliquid DOGE funding cache for CARRY+BASIS; DOGE daily bars from validation.common; _single_coin_decide on DOGE gated by BTC ROC60, validated in reports/universe_generalization.md used to compute the DOGE-gate60 target.
Diagnostics: corr(carry_z, basis_z) = +0.776 | fraction of long days with both signals stretched long (0.5x bucket): 0.6% | fraction with both signals stretched short (1.25x bucket): 0.9%.
| Variant | Days | LongDays | Ret | Sharpe | PF | MaxDD |
|---|
| baseline (gate60, no overlay) | 788 | 141 | +62.16% | 0.69 | 1.27 | -52.35% |
| primary overlay (0.5/1.0/1.25) | 788 | 141 | +150.43% | 1.08 | 1.45 | -36.72% |
| Bucket | Days | Share |
|---|
| 1.0x (no signal) | 117 | 83.0% |
| 1.0x (mixed) | 12 | 8.5% |
| 1.25x (both stretched short) | 7 | 5.0% |
| 0.5x (both stretched long) | 5 | 3.5% |
Bootstrap p(overlay beats baseline, full window): 0.000
Bootstrap p(IS < 2025-01-01): 0.001
Bootstrap p(OOS >= 2025-01-01): 0.016
| Segment | Variant | LongDays | Ret | Sharpe | PF |
|---|
| IS | baseline | 74 | +130.22% | 2.25 | 1.80 |
| IS | primary overlay | 74 | +190.35% | 2.77 | 2.11 |
| OOS | baseline | 67 | -29.57% | -0.38 | 0.86 |
| OOS | primary overlay | 67 | -13.75% | -0.08 | 0.97 |
| look (h) | carry z_enter | basis z_enter | LongDays | Ret | Sharpe | PF | MaxDD |
|---|
| 84 | 1.0 | 1.0 | 141 | +170.50% | 1.16 | 1.50 | -36.04% |
| 84 | 1.0 | 1.5 | 141 | +173.57% | 1.17 | 1.50 | -35.86% |
| 84 | 1.0 | 2.0 | 141 | +125.77% | 1.00 | 1.41 | -43.12% |
| 84 | 1.5 | 1.0 | 141 | +142.85% | 1.05 | 1.44 | -40.29% |
| 84 | 1.5 | 1.5 | 141 | +149.77% | 1.08 | 1.45 | -40.12% |
| 84 | 1.5 | 2.0 | 141 | +121.61% | 0.98 | 1.40 | -44.16% |
| 84 | 2.0 | 1.0 | 141 | +129.12% | 1.00 | 1.41 | -43.14% |
| 84 | 2.0 | 1.5 | 141 | +135.65% | 1.03 | 1.42 | -42.97% |
| 84 | 2.0 | 2.0 | 141 | +116.15% | 0.95 | 1.38 | -46.83% |
| 168 | 1.0 | 1.0 | 141 | +196.19% | 1.24 | 1.54 | -33.63% |
| 168 | 1.0 | 1.5 | 141 | +167.66% | 1.15 | 1.49 | -36.03% |
| 168 | 1.0 | 2.0 | 141 | +144.52% | 1.06 | 1.44 | -36.72% |
| 168 | 1.5 | 1.0 | 141 | +174.39% | 1.17 | 1.50 | -36.64% |
| 168 | 1.5 | 1.5 | 141 | +171.50% | 1.16 | 1.49 | -36.64% |
| 168 | 1.5 | 2.0 | 141 | +150.43% | 1.08 | 1.45 | -36.72% |
| 168 | 2.0 | 1.0 | 141 | +148.19% | 1.07 | 1.45 | -40.44% |
| 168 | 2.0 | 1.5 | 141 | +148.19% | 1.07 | 1.45 | -40.44% |
| 168 | 2.0 | 2.0 | 141 | +135.41% | 1.02 | 1.42 | -40.51% |
| 336 | 1.0 | 1.0 | 141 | +215.39% | 1.29 | 1.57 | -30.44% |
| 336 | 1.0 | 1.5 | 141 | +174.18% | 1.17 | 1.50 | -36.65% |
| 336 | 1.0 | 2.0 | 141 | +124.45% | 0.98 | 1.40 | -42.43% |
| 336 | 1.5 | 1.0 | 141 | +143.38% | 1.05 | 1.44 | -43.28% |
| 336 | 1.5 | 1.5 | 141 | +135.26% | 1.02 | 1.42 | -45.17% |
| 336 | 1.5 | 2.0 | 141 | +113.75% | 0.94 | 1.38 | -45.17% |
| 336 | 2.0 | 1.0 | 141 | +133.57% | 1.01 | 1.42 | -45.17% |
| 336 | 2.0 | 1.5 | 141 | +133.57% | 1.01 | 1.42 | -45.17% |
| 336 | 2.0 | 2.0 | 141 | +113.75% | 0.94 | 1.38 | -45.17% |
- IS baseline Sharpe: 2.25, ret +130.22%, maxDD -33.21%
- IS overlay Sharpe: 2.77, ret +190.35%, maxDD -25.40%
- OOS baseline Sharpe: -0.38, ret -29.57%, maxDD -43.63%
- OOS overlay Sharpe: -0.08, ret -13.75%, maxDD -35.36%
- Pre-registered pass: OOS Sharpe +>= 0.10 AND p < 0.05 AND maxDD not +> 5pp
- Result: PASS
- Inert signal risk: if <5% of long days trigger either 0.5x or 1.25x, the overlay is a no-op and the experiment is a wash. Reported as INERT in the verdict.
- Threshold tuning: the proposal pre-registers look=168h, carry_enter=1.5, basis_enter=2.0. The robustness grid is REPORTED, not selected. Picking the IS winner would be the look-ahead bug class that bit R10-13.
- Hyperliquid only: Binance/Bybit/OKX funding are geo-blocked from this IP (451/403). Hyperliquid started in 2023; the cache is the longest available free source. Binance USDT-M funding IS in the metrics cache but is perpetual-rate only, not the premium index; the BASIS leg specifically needs premium.
- Hyperliquid is one venue: cross-venue funding arbitrage (HL vs Binance) is not what this study tests; this study asks whether HL alone is informationally useful as a sizing overlay on a slow-trend signal.
- CARRY and BASIS are correlated: corr reported above. If corr > 0.7, the composite is one-dimensional and the two-signal framing is misleading; the overlay is then a single-signal gate with two thresholds.
- Multipliers are heuristic: 0.5x and 1.25x are the natural ±50% / ±25% asymmetry, not the result of an optimization. An optimization step would re-introduce the threshold-tuning look-ahead; the honest path is to report the asymmetry you chose and let the verdict be what it is.