Bitcoin Futures Basis Guide: Contango, Backwardation & the CME Carry Trade Explained

 

A trade that reliably paid more than 20% a year during the 2021 bull market now yields less than a US Treasury note. According to CoinDesk's August 3 report citing Glassnode, the three-month bitcoin futures basis has paid less than the two-year Treasury yield since February 2026, a stretch that had run 157 days at the time of writing. The only comparable run on record, August 2022 into January 2023, ended at the cycle low.

 

The reason this matters beyond the arbitrage desks is that the basis is not merely a profit line. It is the price gap between spot and futures, and therefore a direct quote on how much the market will pay to hold bitcoin forward. A widening basis means leveraged longs are queuing up. A flat or negative basis means nobody is paying that premium any more. Once the number drops below the risk-free rate, the capital built on it, including the structural pairing of ETF creations with short CME futures, begins to withdraw, and the withdrawal itself presses on price.

 

CoinDesk's bitcoin price page showed BTC near $82,530 on October 9, up 0.18% on the day, after a brief move above $87,000 that was followed by one of the largest profit-taking sessions of 2026. On the other side of the ledger, Treasury yields compiled by Forbes Advisor for October 8 put the three-month bill at 4.23% and the two-year note at 4.75%. The bar that carry trades must clear has moved higher than it was at the start of the year.

 

 

 

The basis is the spot-to-futures price gap, and it only becomes comparable once annualized. Divide the futures price by spot, subtract one, then scale by 365 over days to expiry, and the result can be set directly against a Treasury yield.

 

Contango and backwardation describe the shape of the curve. Futures above spot is contango, the normal state for bitcoin. Futures below spot is backwardation, which typically appears only during panic selling or forced liquidation.

 

Cash and carry is not risk-free. Buying spot and shorting an equal amount of futures locks in convergence at expiry, but margin calls, roll costs, fees on both legs and counterparty exposure all eat into the printed number.

 

The spread has fallen under the hurdle rate. CoinDesk's early-August figure put carry near 3% against roughly 3.8% on two-year Treasuries, and the Fed's September hike has lifted the risk-free benchmark since.

 

Capital has already voted. crypto.news reported on April 9 that average daily open interest in CME bitcoin futures fell to about $7.2 billion in early April, the lowest since February 2024.

 

Two sources can print basis figures that differ by a factor of two. Tenor, timestamp, venue and whether costs are netted all change the answer, so reading the methodology matters more than memorizing the number.

 

 

 

The Federal Reserve confirmed in its implementation note for September 16 that the target range moved to 3.75% to 4%, with interest on reserve balances at 3.90%. For arbitrage capital, that figure is the opportunity cost. A capital-intensive basis trade that returns less than a short-dated Treasury has no reason to exist, regardless of where bitcoin trades.

 

The hurdle rose while the basis did not. Per TFTC's summary of the Glassnode weekly report on August 3, the three-month futures basis has sat below the two-year Treasury yield every month since February, only the second such stretch on record. The first, from August 2022 to January 2023, lasted roughly five months and ended at the previous cycle's low. The analogue is not a forecast, but it does indicate that prolonged sub-Treasury carry tends to coincide with the trough of risk appetite rather than the middle of a rally.

 

 

Retreating arbitrage capital leaves fingerprints in positioning data. According to CME positioning cited by Yahoo Finance on August 11, leveraged funds held 4,243 long against 11,483 short standard bitcoin contracts as of August 4, a net short of 7,240 contracts, or roughly 36,200 BTC at five coins per contract, while micro contracts showed a net long of about 3,943 contracts, or 394 BTC. The direction of travel is what counts: that net short has shrunk by roughly half in BTC terms over the past year.

 

Those shorts were never bearish bets. They were the hedge leg of an arbitrage structure. CoinDesk reported on August 10 that CryptoQuant founder Ki Young Ju argued a traditional carry trade cannot be run from an aggregate net-long futures position, while noting that part of the shift may reflect basis traders closing shorts rather than placing new bullish bets. The distinction decides whether the data reads as a sentiment turn or a structural exit.

 

 

 

In raw form the basis is just a spread. Suppose spot sits at $82,530 and the three-month contract trades at $83,500. The gap is $970, or about 1.18%. On its own that means little, because it covers 90 days rather than a year. Annualized, 1.18% scaled by 365 over 90 gives roughly 4.8%, which is the figure that belongs next to a 4.23% three-month bill. These numbers illustrate the arithmetic and do not represent live quotes.

 

Note that annualizing a near-dated contract amplifies noise. A 0.1% spread on a contract three days from expiry annualizes to 12%, yet nobody captures that, because execution costs consume almost all of it. This is why the three-month tenor has become the institutional reference.

 

 

Futures above spot is contango, the default state in bitcoin, reflecting carrying costs and leverage demand together: more participants want long exposure than want to provide it, so longs pay a premium to hold forward. Futures below spot is backwardation, which means the market is discounting future bitcoin, and it tends to appear only in sharp declines, forced liquidation or liquidity droughts. The crypto.news report cited above notes that the CME-to-spot basis has turned negative during some stress episodes.

 

The slope carries information too. A front month richer than deferred months usually signals short-term speculative heat, while deferred months trading above the front end looks more like an ordinary cost-of-carry structure.

 

 

Within the same week, CoinDesk's August 3 report put carry near 3%, while CryptoSlate's matched-date calculation published August 11 showed the August contract at 7.89% annualized, September at 6.25% and December at 5.69% using CME settlements for August 7, against a two-year Treasury of 4.19% that day. The two are not in conflict. The latter is a gross figure before funding, margin, fees and execution costs on either leg, and it uses near-dated contracts, while the former uses a three-month rolling measure.

 

The official methodology is worth knowing. Per CME's cryptocurrency BasisWatch and implied rate tool, spot comes from the CME CF Spot-Quoted Cryptocurrency Marker, a 60-second time-weighted average price from 3:59 to 4:00 p.m. ET, paired with the nearest monthly contract, rolling on the last Tuesday of the maturity month. Glassnode's three-month annualized basis chart applies a different convention. Checking which one a figure comes from beats arguing over the second decimal.

 

 

 

The standard structure buys spot bitcoin or a spot ETF and sells an equivalent amount of CME futures. Directional risk is hedged away and the spread is what remains. Because futures must converge to spot at expiry, holding to settlement theoretically captures the spread locked in at inception, whatever bitcoin does in between.

 

Contract mechanics decide whether the hedge is precise. Per CME's cryptocurrency futures FAQ, the standard bitcoin contract covers 5 BTC and the micro contract 0.1 BTC, both cash-settled, with final settlement referencing the CME CF Bitcoin Reference Rate published at 4:00 p.m. London time and daily settlement using the volume-weighted average of trades between 2:59 and 3:00 p.m. CT. Cash settlement removes wallets and on-chain transfers from the workflow, which is a large part of why traditional managers use CME rather than offshore venues. For desks that need to trade the spread at the index close, CME also offers Basis Trade at Index Close.

 

 

Several layers sit between the headline yield and realized return. The spot leg either ties up capital or needs financing, and financing comes straight out of the spread. An ETF used as the long leg adds a management fee. The futures leg requires maintenance margin, which consumes capital and can be called in volatile conditions. Both legs incur fees and slippage, and holding across quarters adds roll cost at whatever basis prevails on the roll date.

 

That is why CryptoSlate flagged its 7.89% as a gross figure and warned that funding, margin, fees and execution can erase the spread. On a fully funded trade printing 5% gross, keeping two to three points net is a respectable outcome, while the three-month bill currently pays 4.23% with no margin calls attached.

 

 

Convergence is certain; the path is not. Before expiry, if the basis widens instead of narrowing, the futures leg posts a mark-to-market loss ahead of the spot leg's gain, and margin pressure follows. Large positions in that situation are often cut by risk managers, and the act of cutting moves the market. Exchange and custodian counterparty risk and tracking error between the two legs round out the list.

 

 

 

The clearest evidence of retreat sits in activity data. Per the crypto.news report, average daily open interest in CME bitcoin futures fell below $8 billion in March and reached about $7.2 billion in early April, the lowest since February 2024 and the extension of a five-month decline, while March volume was roughly half the January 2025 peak on the same measure. The same piece cites a Binance research note from January summarizing the shift as the end of the arbitrage era, with Wall Street stepping back from the bitcoin basis.

 

Where that activity went matters. Liquidity has rotated toward offshore venues and perpetual swaps, which moves part of price discovery with it. Perpetuals are anchored by funding rates rather than convergence to expiry, so the two instruments measure different kinds of leverage demand, a distinction set out in our explainer on funding rates and open interest.

 

 

For much of the past two years, a meaningful share of spot ETF creations was not directional allocation but the long leg of an arbitrage structure. That changes how flow data should be read. When the basis pays, net inflows overstate genuine buying interest because hedged positions are mixed in. When the basis collapses, unwinding shows up as net outflows that have nothing to do with a bearish view. CryptoSlate's caveat applies here too: flow and positioning data show aggregates and cannot tie a specific ETF buyer to a specific futures hedge.

 

Cross-checking is the safer habit, using the bitcoin ETF flow tracker alongside the structural differences covered in bitcoin versus a bitcoin ETF.

 

Whether the curve steepens or flattens, it lands in the spot price. Open the BTC spot market and read it against the futures curve

 

 

October 2025 provides the complete case study. According to Amberdata's reconstruction of that unwind, the 30-day annualized basis peaked at 15.3% during the policy-euphoria phase of January 2025, stood at 6.9% before the October break, troughed at 4.5% during the crash and settled at 5.2% afterwards. The same analysis calculates that, measured as excess return over T-bills, the trade was genuinely worth running on only about 31 days across 2025.

 

The feedback loop is mechanical. In a sell-off, spot falls faster than futures, so a supposedly hedged book posts a net loss, risk managers force reductions, and closing the trade means selling spot and buying back futures. The spot selling adds directly to the decline, which deepens the compression that started it. Amberdata also notes that part of October's ETF outflows were arbitrage unwinds rather than investor exits.

 

This explains something that confuses many observers: how a market-neutral structure becomes an amplifier at the worst moment. Hedging removes directional risk, not basis risk, and basis risk peaks precisely when everyone reaches for the exit at once.

 

 

 

Most retail participants will never run the trade, because financing costs and operational requirements make it impractical. The indicator, however, is free. A rising annualized basis with deferred months bid signals that leveraged longs are stacking up, and pullbacks in that environment tend to be violent. A basis pinned to or below the risk-free rate signals that speculative demand has already cleared out, which usually caps downside but raises the time cost of waiting.

 

Cross-verification is the better use. The basis reflects dated leverage demand while funding rates reflect immediate imbalance in perpetuals, and only when both heat up at once is the signal convincing. When funding spikes while the basis stays flat, as in the episode CoinDesk documented on October 2, with open interest rising about 27,000 BTC to roughly 653,000 BTC, perpetual funding climbing from around 3% to 10% and price moving from $83,500 to $86,500, the move looks more like short-term leverage than durable demand.

 

 

The FOMC meets next on October 27 and 28, and per the Fed's meeting calendar that is the next reset of the hurdle rate. If yields keep rising while the basis stays flat, the return of arbitrage capital gets pushed further out. The basis crossover is the other marker: three-month carry rising back above the two-year Treasury would be the first such reading since February, and it would say more about institutional demand than a price breakout does. Whether CME open interest recovers from the $7.2 billion low is the confirming data point.

 

 

If rates fall back, with the Fed pausing or easing, the hurdle drops while leverage demand recovers, carry clears Treasuries again, arbitrage capital returns, and the hedged portion of ETF creations comes back with it.

 

If current conditions persist, the basis keeps trading at or below the risk-free rate, CME positioning stays depressed, and price is driven by spot supply and demand rather than leverage, which lowers volatility while also weakening directional conviction.

 

Under stress, a rapid decline compresses the basis further or flips it into backwardation, surviving carry positions are liquidated, and the spot selling amplifies the drop along the October 2025 path. The low-probability tail is a failure at a major custodian or clearing venue, at which point every basis calculation becomes irrelevant and counterparty quality is the only question that matters.

 

In all three, remember that the basis describes market structure rather than timing. It tells you which side the leverage sits on, not where price goes tomorrow.

 

 

For James Mitchell, the most underrated implication of this compression is that pricing power in bitcoin is passing from leverage back to spot. Carry above 20% in 2021 was rent collected on market inefficiency, the product of too little capital willing to take the other side against too many participants wanting leveraged length. The disappearance of that rent looks like a revenue problem for arbitrage desks, but it is really this market's financing cost converging toward that of conventional assets.

 

Two misreadings look likely. The first treats a depressed basis as a bearish signal. The basis prices leverage demand, not spot demand, and a market with flat carry and steady spot bids is less volatile rather than more dangerous. The second overlooks the hedged component inside ETF flows. A good deal of the institutional-adoption narrative of the past two years was counting the long leg of an arbitrage structure, and when carry fell below Treasuries that capital left for arithmetic reasons. CryptoSlate's point that no public dataset ties an individual creation to an individual hedge is exactly why this attribution needs care.

 

The variable worth tracking from here is the spread between three-month carry and the three-month bill, not the level of the basis itself. At 4.23%, the risk-free alternative is a real competitor, and only a sustained positive spread would show that leveraged demand has genuinely returned. Glassnode's 157-day record and the five-month precedent from 2022 offer a distribution to reference rather than a prediction, though they do establish that this state has historically been finite. Pairing that with the pace of any recovery in CME open interest from the $7.2 billion low separates a durable return from a pulse.

 

The cross-asset lesson is that crypto is retracing a familiar path. Equity index futures once carried wide bases too, and as market-making improved and participation broadened, the spread converged toward financing cost, excess profit disappeared, and the market functioned better for it. Bitcoin has reached the same junction. The cost is that speculative premium is no longer generous. The benefit is that price discovery now leans on real supply and demand. For long-term holders that is not bad news, but for capital that treated a fat basis as permanent, the business model needs rewriting.

 

 

 

The basis is the difference between the futures price and the spot price. Dividing that gap by spot and annualizing it by days to expiry produces a figure comparable to a bond yield. With spot at $82,530 and a three-month contract at $83,500, the gap is about 1.18%, or roughly 4.8% annualized. It measures what the market will pay to hold bitcoin forward, which is why it doubles as a thermometer for leverage demand.

 

 

Futures trading above spot is contango, the normal condition in bitcoin, reflecting longs paying a premium for forward exposure. Futures trading below spot is backwardation, meaning the market is discounting future bitcoin, which typically appears only during panic selling, forced liquidation or a liquidity drought. The slope matters as well, since a front month richer than deferred contracts usually indicates short-term speculative heat.

 

 

No. Buying spot and shorting futures does neutralize directional risk, and convergence at expiry is certain, but margin calls, roll costs, fees and slippage on both legs, and exchange and custodian counterparty exposure all apply. The central hazard is basis risk: if the spread widens during the holding period, the futures leg takes a mark-to-market loss first, which can force an early exit and turn a locked-in gain into a realized loss.

 

 

Because the hurdle rate rose while the basis did not. CoinDesk's early-August figure put carry near 3% against roughly 3.8% on two-year Treasuries, and Glassnode's data shows three-month carry below the two-year yield since February. After the Fed lifted the target range to 3.75% to 4% in September, the three-month bill stood at 4.23% on October 8. Tying up capital for a single-digit return that also requires maintenance margin no longer makes economic sense.

 

 

Because the conventions differ. CME's BasisWatch pairs a 60-second time-weighted spot average ending at 4:00 p.m. ET with the nearest monthly contract, Glassnode uses a three-month rolling measure, and some outlets calculate contract by contract from daily settlements. Some figures are gross, before financing, margin and trading costs. Seeing 3% and 7.89% in the same week comes almost entirely from tenor selection and whether costs are netted out.

 

 

It describes structure better than it times turns. An elevated basis usually means leveraged longs are crowded, which makes pullbacks more prone to cascading. A basis at or below the risk-free rate means speculative demand has already cleared. The available historical reference is that the August 2022 to January 2023 stretch of sub-Treasury carry ended at the cycle low, but one sample is not a rule.

 

 

Yes. The basis is public, and reading it requires no participation. The practical method is to view it alongside perpetual funding, open interest and ETF flows: basis and funding rising together indicate real leverage accumulation, while funding alone spiking points to short-term speculation. For spot-focused investors, the data mainly informs when to reduce leverage or slow accumulation, and the mechanics of getting exposure are covered in the guide to buying bitcoin and the BTC purchase walkthrough.

 

 

The information above is provided for general market information and analysis only and does not constitute investment advice, financial advice, legal advice, tax advice or a recommendation to trade. Prices of crypto assets, derivatives and other related financial instruments can fluctuate sharply, and leveraged or arbitrage structures can produce losses beyond expectations under certain market conditions. Past performance, technical indicators and on-chain data do not guarantee future results. The prices, basis levels, positioning data, yields and research conclusions cited here reflect publicly available information at the time of publication and may change at any time, so the latest disclosures from the relevant exchanges, data platforms and regulators should be treated as authoritative. Readers should conduct their own research and make decisions based on their own financial circumstances, investment objectives and risk tolerance, consulting a qualified professional where appropriate. The MEXC Crypto Pulse team accepts no liability for any direct or indirect loss arising from the use of this information.

 

 

James Mitchell specializes in technical analysis, market trends, and trading strategies for both Bitcoin and altcoins. Based in London, he has over 10 years of experience in financial markets. Before joining MEXC Learn, James worked as a senior analyst at a leading European investment firm, where he developed expertise in risk management and quantitative trading. His transition to cryptocurrency markets began in 2017, and he has since become recognized for his data-driven approach. He holds a Master's degree in Financial Economics from the London School of Economics. His analytical approach combines traditional technical analysis with on-chain metrics to provide readers with actionable insights.

 

Areas of Expertise: Technical Analysis, Market Trends and Cycles, Trading Strategies, Bitcoin and Altcoin Analysis, Risk Management.

 

 

 

Want the fastest access to MEXC's latest updates? Join our official Telegram group now!

MEXC Platform: Download App | Zero Fee Trading | Buy USDT Now

Join MEXC Community: X (Twitter) | Telegram | Discord

Get Insights: MEXC Learn | Terms of Service | MEXC Blog | MEXC News

Account Verification: Understand KYC | How to Complete KYC

External Content Platforms: Substack | Medium | Paragraph | LinkedIn | X(News)