Funding rate arbitrage on Hyperliquid
Verified against Hyperliquid docs: Funding and Hyperliquid docs: Margining · by Hyperliquid Academy
The structure
Two positions, opposite directions, same size:
Long spot. You own the asset. Its price can do anything.
Short the perpetual. Same notional. It loses what the spot leg gains, and gains what the spot leg loses.
Net price exposure: roughly zero. What remains is the funding on the short leg, which is paid to you every hour while the rate is positive.
That is the whole idea, and it is a genuinely old one. Its appeal here is that both legs sit in one account on one venue, so there is no transfer between exchanges and no delay when you need to unwind.
Why the rate is usually positive
Funding exists to hold the perpetual price near the spot price. When the perpetual trades above spot — which happens when positioning leans long, which is most of the time in crypto — longs pay shorts.
So the short side of the trade is, on average, the paid side. “On average” is doing real work in that sentence.
| Market | Price | 24h | 24h volume | Open interest | Funding / 1h | Max leverage |
|---|---|---|---|---|---|---|
| BTC | $75,983.60 | +0.36% | $3.16B | $2.91B | 0.00082% | 40x |
| ETH | $2,402.30 | +0.08% | $1.4B | $2.36B | 0.00125% | 25x |
| ZEC | $1,310.40 | +17.21% | $866.23M | $812.81M | 0.00125% | 10x |
| HYPE | $78.025 | +1.46% | $518.05M | $1.61B | 0.00125% | 10x |
| SOL | $98.215 | +1.30% | $173.83M | $521.84M | 0.0012% | 20x |
| XRP | $1.289 | +0.27% | $106.7M | $190.02M | -0.0012% | 20x |
| ARB | $0.16543 | +14.77% | $58.16M | $26.85M | 0.00125% | 10x |
| NEAR | $2.565 | +10.57% | $49.07M | $161.29M | 0.00125% | 10x |
| LIT | $4.518 | +9.66% | $44.23M | $190.87M | 0.00125% | 5x |
| PUMP | $0.003611 | +2.06% | $42.99M | $136.93M | -0.00229% | 10x |
Read that table before assuming the trade is available. A rate near zero does not clear the costs, and a negative one means you are paying to hold a position with no price exposure, which is the worst of both.
Where it actually breaks
The costs are the strategy
The idea is trivial. Whether it makes money is entirely an arithmetic question, and the arithmetic has four terms people routinely leave out.
Spot fees, twice. You buy the spot leg and eventually sell it. Spot rates are higher than perps rates at every tier. The spot schedule.
Perp fees, twice. Opening and closing the short.
Slippage on four executions. Two legs in, two legs out. On anything outside the deepest markets this can exceed all the fees combined. Judging depth first.
The rate changing. This is the one that decides most outcomes. The rate you entered on is not a contract; it is a snapshot. It can halve, or flip, in the hours after you open.
Add them up and a clear pattern emerges: the trade needs the rate to persist, not merely to be attractive right now. A position that pays for two hours and then costs you for six was a losing trade with a correct thesis.
The risks that are not price risk
Liquidation of the short leg. The perpetual short still has a liquidation price, and it sits above your entry. If the market rallies hard and the short is thinly margined, it can be liquidated even though your spot leg gained exactly as much. You then hold a long-only position in a rally you had deliberately hedged out of — which is not a disaster, but it is not the trade you put on. Margin the short generously and treat that margin as part of the capital cost.
Capital efficiency. You fund the spot purchase in full and margin the short on top. The return on the funding is calculated against all of it, which is why the headline rate flatters the strategy.
Unwind risk. Closing both legs at once, in a market that is moving, is the moment the execution costs land. Unwinding one leg first leaves you directional for as long as it takes to do the other.
Concentration. Doing this on one market means one funding regime, one book and one liquidation price. Doing it on several thin markets to diversify usually just multiplies the slippage.
When it is worth doing
The rate is strongly positive and has been for a while. Persistence is the variable that matters most, and the recent history is the only evidence available.
The market is deep. Four executions on a thin book is how a positive-carry trade becomes a negative one before the first funding payment.
Your capital has no better use. The comparison is not against zero. It is against what the same capital would earn elsewhere, including HLP, at a considerably lower operational burden.
You can monitor the rate. A trade that depends on an hourly variable needs someone or something watching it hourly. This is why automation suits it.
Sizing it honestly
Use the funding calculator with the live rate rather than an assumed one, and then subtract the four cost terms above before deciding anything.
Do that once and the usual conclusion arrives on its own: the trade is worth doing at size, on deep markets, when the rate is unusually high — and is not worth doing at small size on an ordinary rate, because the round trip eats the carry.
That is not an argument against the strategy. It is the strategy: it is a spread business, and spread businesses live or die on execution cost.
Where to go next
How the rate is actually computed, including the clamp and the cap, and the mechanics of the short leg, which is the half of this trade with a liquidation price attached.
Frequently asked questions
What returns are realistic?
Whatever the funding rate is, minus your costs, for as long as the rate holds. That is not a number anyone can promise: the rate moves hourly, it flips sign, and a strategy quoted as an annualised figure from a good week is describing that week, not a yield.
Does it carry no risk at all?
It carries plenty. The price exposure cancels; the rest does not. The rate can turn negative, the short leg can be liquidated if its margin is thin, and both legs cost fees to open and close whatever happens afterwards.
Why do both legs on Hyperliquid?
One account, one margin system, no transfer between venues and no withdrawal delay if you have to unwind quickly. Splitting the legs across two venues adds a transfer that takes exactly as long as you cannot afford to wait.
How much capital does it take?
More than a directional trade of the same size, because you fund a spot purchase and margin a short at once. That is the real constraint, and it is why the annualised figures look better than the return on capital deployed.
When is the trade worth doing?
When the rate is strongly positive, expected to persist for long enough to clear the round-trip costs, and the market is deep enough to enter and exit both legs without paying it back in slippage.
Can I automate it?
It is one of the better-suited strategies here, because funding settles hourly and both legs live in one account. The hard part is the unwind logic, not the entry.
Sources
- Hyperliquid docs: Fundinghyperliquid.gitbook.io
- Hyperliquid docs: Margininghyperliquid.gitbook.io
We link the primary source for every number on this page. If a figure here disagrees with the Hyperliquid documentation, the documentation is right and we want to know.