Funding carry: 16 Backtests
Data last updated
Rules that use the funding payments on crypto perpetual futures as a signal.
1 of 16 strategies finished ahead of the benchmark on the raw number, and 0 had an unadjusted p-value under 0.05. None cleared the correction for trying 1,774 tests in total (a q-value below 0.10), so no result here is a confirmed edge. Scored 2026-10-31, using data up to 2025-03-15.
Strategies here tested on crypto, ETFs. See the full family index, the full method and every result, or download the raw registry (JSON).
Results
| Rule | Market | Vs benchmark | Sample | p-value | Corrected q | Details |
|---|---|---|---|---|---|---|
| Treasury duration chosen by the 2s10s slope level Source: Fama & Bliss 1987 AER (forward spreads predict bond returns); Campbell & Shiller 1991 RES | ETFs | +1.6 pts/yr | 102 rebalances | 0.150 | 1.000 | On-site receipt |
| Crypto Funding Exit Source: Schmeling, Schrimpf & Todorov 2023 | crypto | −12.4 pts/yr | 155 trades | 1.000 | 1.000 | On-site receipt |
| Terra6 Price Sign Twin Source: none published (Tickfloor research-desk test) | crypto | −25.9 pts/yr | 2,832 trades | 1.000 | 1.000 | On-site receipt |
| Trend with a funding budget exit: leave once cumulative funding paid since entry exceeds 2% Source: Tickfloor research batch: crypto ideas | crypto | −26.2 pts/yr | 1,175 trades | 1.000 | 1.000 | On-site receipt |
| Terra6 Funding Sign Carry Source: none published (Tickfloor research-desk test) | crypto | −37.5 pts/yr | 594 trades | 1.000 | 1.000 | On-site receipt |
| Buy-after-a-drop matched control (3d return decile, funding-gated sample) Source: none published (Tickfloor research-desk test) | crypto | −50.3 pts/yr | 1,336 trades | 1.000 | 1.000 | On-site receipt |
| Goldilocks funding regime: long only while 30-day funding is positive but not euphoric Source: Tickfloor research batch: crypto ideas | crypto | −51.9 pts/yr | 723 trades | 1.000 | 1.000 | On-site receipt |
| Funding slope: 7-day mean crossing above 30-day mean from below zero Source: Tickfloor research batch: crypto ideas | crypto | −53.1 pts/yr | 987 trades | 1.000 | 1.000 | On-site receipt |
| Return-orthogonalised funding residual, bottom decile Source: Own hypothesis: the residual of perp funding after removing its trailing-return componentFull test: Funding Rate and Open Interest | crypto | −56.4 pts/yr | 1,232 trades | 1.000 | 1.000 | On-site receipt |
| Paid-to-hold trend: long while weekly funding is negative and price above 20-day mean Source: Tickfloor research batch: crypto derivatives | crypto | −56.7 pts/yr | 454 trades | 1.000 | 1.000 | On-site receipt |
| Unlevered rally: 10-day gain while funding fell, long continuation Source: Tickfloor research batch: crypto derivatives | crypto | −57.1 pts/yr | 392 trades | 1.000 | 1.000 | On-site receipt |
| Funding persistence: sustained premium streak as trend confirmation Source: Tickfloor research batch: crypto ideas | crypto | −57.6 pts/yr | 171 trades | 1.000 | 1.000 | On-site receipt |
| Trend long gated by funding crowding percentile (own 180-day history) Source: Tickfloor research batch: crypto derivatives | crypto | −62.1 pts/yr | 766 trades | 1.000 | 1.000 | On-site receipt |
| Negative funding percentile dip inside an uptrend (shorts pay while trend intact) Source: Tickfloor research batch: crypto ideas | crypto | −62.7 pts/yr | 232 trades | 1.000 | 1.000 | On-site receipt |
| Shorts building into negative funding: squeeze setup Source: Tickfloor research batch: crypto ideas | crypto | −82.2 pts/yr | 158 trades | 1.000 | 1.000 | On-site receipt |
| Funding + OI combined squeeze detector Source: Coinglass/Velo Data squeeze commentaryFull test: Funding Rate and Open Interest | crypto | −83.2 pts/yr | 34 trades | 1.000 | 1.000 | On-site receipt |
- Treasury duration chosen by the 2s10s slope level+1.6 pts/yr
Source, p-value and result file
- Source
- Fama & Bliss 1987 AER (forward spreads predict bond returns); Campbell & Shiller 1991 RES
- p-value
- 0.150
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Crypto Funding Exit−12.4 pts/yr
Source, p-value and result file
- Source
- Schmeling, Schrimpf & Todorov 2023
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Terra6 Price Sign Twin−25.9 pts/yr
Source, p-value and result file
- Source
- none published (Tickfloor research-desk test)
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Trend with a funding budget exit: leave once cumulative funding paid since entry exceeds 2%−26.2 pts/yr
Source, p-value and result file
- Source
- Tickfloor research batch: crypto ideas
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Terra6 Funding Sign Carry−37.5 pts/yr
Source, p-value and result file
- Source
- none published (Tickfloor research-desk test)
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Buy-after-a-drop matched control (3d return decile, funding-gated sample)−50.3 pts/yr
Source, p-value and result file
- Source
- none published (Tickfloor research-desk test)
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Goldilocks funding regime: long only while 30-day funding is positive but not euphoric−51.9 pts/yr
Source, p-value and result file
- Source
- Tickfloor research batch: crypto ideas
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Funding slope: 7-day mean crossing above 30-day mean from below zero−53.1 pts/yr
Source, p-value and result file
- Source
- Tickfloor research batch: crypto ideas
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Return-orthogonalised funding residual, bottom decile−56.4 pts/yr
Full test: Funding Rate and Open Interest
Source, p-value and result file
- Source
- Own hypothesis: the residual of perp funding after removing its trailing-return component
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Paid-to-hold trend: long while weekly funding is negative and price above 20-day mean−56.7 pts/yr
Source, p-value and result file
- Source
- Tickfloor research batch: crypto derivatives
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Unlevered rally: 10-day gain while funding fell, long continuation−57.1 pts/yr
Source, p-value and result file
- Source
- Tickfloor research batch: crypto derivatives
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Funding persistence: sustained premium streak as trend confirmation−57.6 pts/yr
Source, p-value and result file
- Source
- Tickfloor research batch: crypto ideas
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Trend long gated by funding crowding percentile (own 180-day history)−62.1 pts/yr
Source, p-value and result file
- Source
- Tickfloor research batch: crypto derivatives
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Negative funding percentile dip inside an uptrend (shorts pay while trend intact)−62.7 pts/yr
Source, p-value and result file
- Source
- Tickfloor research batch: crypto ideas
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Shorts building into negative funding: squeeze setup−82.2 pts/yr
Source, p-value and result file
- Source
- Tickfloor research batch: crypto ideas
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Funding + OI combined squeeze detector−83.2 pts/yr
Full test: Funding Rate and Open Interest
Source, p-value and result file
- Source
- Coinglass/Velo Data squeeze commentary
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
How to read these results
"Vs benchmark" is the strategy's annualized return minus an equal-weight benchmark, rebalanced daily, on the same assets and days. Strategies are charged trading costs; the benchmark pays none. Ahead is not a confirmed edge: the p-value is the unadjusted chance of a gap this large by luck, and the corrected q allows for how many strategies were tried, because testing hundreds of ideas throws up lucky winners. Every strategy saw only data up to the same cutoff date. The receipt is the frozen result file for that rule.
"Sample" is trades for rules that enter and exit a position, and rebalances for portfolio rules. A rebalance is a scheduled date on which a portfolio rule re-ranks and held at least one position. It is the sample size for portfolio rules. A holdings change is a rebalance where what the portfolio held, or its weights, differed from the previous one. Only these cost money to trade. The two are different counts and are never swapped. Definitions.