For intermediate traders, this guide shows how to run a spot-perp crypto carry trade with real funding math, sizing rules, exits and failure points without betting on direction.
Crypto carry trade strategy works when the perp market pays enough funding to make a hedged spot position worth the operational risk. This is for traders who already understand perps and want rules for buying spot, shorting perps, collecting funding, and knowing when to get flat.
The clean version is simple: own the coin in spot, short the same notional in perpetual futures, and let positive funding flow from longs to shorts. The hard part is not the hedge; it is avoiding crowded funding, thin margin, and exchange friction.
I treat positive funding as raw material, not a trade by itself. On Binance BTCUSDT or Bybit ETHUSDT, 0.01% per 8 hours is often too small after fees, slippage, and the chance that funding flips before the next settlement.
| Funding rate | Simple annualized | Action |
|---|---|---|
| 0.01% per 8h | 10.95% | Usually skip unless fees are near zero |
| 0.03% per 8h | 32.85% | Tradable on BTC or ETH if books are deep |
| 0.05% per 8h | 54.75% | Strong candidate after fee check |
| 0.10% per 8h | 109.5% | Crowded; size down and plan the exit first |
My entry filter: funding is at least 0.03% per 8h on BTC or ETH, or 0.05% per 8h on liquid alts; perp premium is under 0.25%; and total round-trip fees are less than one day of expected funding. Binance, Bybit, and OKX commonly use 8-hour funding on major perps, while Coinbase International uses hourly funding, so normalize every quote to a daily rate before comparing venues.
VoiceOfChain tracks funding spreads and exchange divergence in real time across Binance, Bybit and OKX, so you can see live spot-perp carry candidates without building your own monitor. voiceofchain.com
The classic positive carry setup is long spot and short the same notional perp. If BTC trades at $60,000 and you want $10,000 notional, buy 0.1667 BTC spot on Coinbase or OKX and short about $10,000 of BTCUSDT perpetuals on Bybit, Binance, or OKX.
| Step | Action | Reason |
|---|---|---|
| 1 | Buy $10,000 BTC spot on OKX | Creates the long asset leg |
| 2 | Short $10,000 BTCUSDT perp on Bybit | Neutralizes directional BTC exposure |
| 3 | Confirm funding is positive before settlement | Longs should be paying shorts |
| 4 | Record entry basis and fees | You need this for the exit math |
When funding is positive, long perp holders pay short perp holders; when it is negative, shorts pay longs. If the rate flips negative before settlement, the carry trade becomes a cost center immediately.
I do the math in dollars first, APY second. APY screenshots are seductive, but the trade lives or dies on actual funding collected versus fees, basis movement, and margin stress.
| Item | Amount |
|---|---|
| Account equity | $20,000 |
| Spot BTC bought | $10,000 |
| Perp short notional | $10,000 |
| Perp margin posted | $5,000 |
| Cash buffer | $5,000 |
| Funding rate | 0.05% per 8h |
| Expected funding | $5 per settlement, $15 per day, about $450 per 30 days |
That setup earns about 4.5% per month on the $10,000 notional before fees, or 2.25% per month on the full $20,000 account. If your spot and perp round-trip cost is 0.16%, the trade costs $16 per $10,000 notional, so 0.05% per 8h funding needs a little over one day to break even.
At 0.01% per 8h, the same $10,000 position earns only $3 per day. That means the same $16 fee drag needs more than five days of stable funding, which is usually not enough edge for the operational risk.
A carry trade stop is not a normal BTC price stop because the book should be close to delta neutral. I use basis stops, funding stops, and margin stops; the goal is to stop the hedge from becoming a leveraged directional bet.
| Stop type | Rule | Why it matters |
|---|---|---|
| Basis stop | Exit if perp premium widens 0.50% against entry and next 3 funding payments do not cover it | Prevents paying too much mark-to-market for future yield |
| Funding stop | Exit if funding drops below 0.01% for 2 settlements or flips negative | The edge is gone |
| Margin stop | Reduce if liquidation buffer falls below 20% on BTC or ETH, 30% on alts | Mark-price spikes can liquidate the short leg |
| Account stop | Cut risk if total carry-book drawdown hits 1.5% of account equity | Keeps one crowded trade from damaging the portfolio |
The common mistake is using high leverage because the hedge looks neutral. During a liquidation cascade, the perp mark can detach from spot, funding can flip, and the short can get stressed while the spot leg is sitting on another exchange.
Trader's risk caveat: carry fails when the trade becomes crowded, not when it looks boring. When open interest is rising faster than price and longs are paying 0.10%-0.30% per 8h, smaller size beats chasing the headline APY.
The key takeaway: carry is not free yield; it is paid operational work plus liquidity risk. Use it only when funding is high enough after fees, hedge precisely, size small enough to survive mark-price stress, and exit when funding stops compensating you. On BTC and ETH, a boring 0.03%-0.05% per 8h setup with deep books often beats chasing a 0.30% altcoin print. Treat every funding payment as earned only after the settlement hits your account.