BitMart users and employees have demanded that the crypto exchange disclose its reserves, explain reported withdrawal restrictions, and publish a repayment plan by Aug. 19 as questions continue over customer funds and unpaid staff compensation.

A public statement posted on X and addressed to BitMart founder Sheldon Lee and Yi Li said a large number of users remained unable to withdraw their assets, while some employees had not received their previous month’s salary or compensation owed to them.

The statement called for BitMart to provide verifiable information on its wallets, assets, liabilities and reserves available to meet customer withdrawals. It also asked the exchange to accept independent third-party scrutiny of the figures instead of relying on company statements about its financial position.

In a translated version of the statement, the group called on BitMart to “disclose the wallets,” “disclose the assets,” “disclose the liabilities” and “disclose the actual usable reserves.”

The demands come weeks after BitMart announced plans to wind down its trading platform after nine years of operation. On July 26, the exchange said it would stop new registrations, deposits and new trading activity while allowing customers to withdraw funds during the shutdown process.

BitMart said at the time that all trading services would end on Aug. 26 before the company ceased operations on Jan. 31, 2027. The exchange attributed the decision to its operating conditions, market environment, and future strategy without identifying a specific financial or operational event behind the closure.

BitMart users seek answers over withdrawal restrictions

Alongside the request for reserve information, the open letter asked BitMart to explain why users were reportedly still unable to complete withdrawals normally and when management first became aware of the problems.

The statement asked who decided to restrict withdrawals, when the decision was made, and whether BitMart continued encouraging customers to deposit, trade or leave assets on the platform after management became aware of withdrawal or funding issues.

BitMart had addressed separate withdrawal complaints before announcing its shutdown. In a statement published in June, the exchange said reports of users being unable to withdraw or facing account restrictions were mainly connected to risk controls targeting what it described as an organized scheme designed to exploit platform activity subsidies.

Following the July shutdown announcement, BitMart said withdrawals would remain available but could be subject to additional checks involving identity verification, devices, IP addresses, withdrawal destinations, sources of funds and sanctions screening. The exchange also warned that a large number of requests could increase processing times.

The current open letter disputed whether the shutdown notice alone addressed the problems reported by users and employees, arguing that customers needed an accounting of the assets available to meet their balances.

Such disclosures would require information beyond a list of wallet holdings to establish a complete picture of an exchange’s finances. A proof-of-reserves report can verify crypto held in identified wallets against customer balances at a particular time, but it does not necessarily reveal off-chain liabilities, borrowed assets or other obligations.

Recent exchange disclosures show how those figures can be presented. In July, crypto.news reported that Binance reserve data showed customer Bitcoin holdings increasing by 7,715 BTC during June, with the report based on a July 1 snapshot. The same report noted that reserve snapshots do not constitute complete financial audits.

Open letter calls for investigation of related accounts

The BitMart statement also sought an investigation into accounts, affiliated companies, trusts and other arrangements that may have handled funds connected to users.

Yi Li was specifically asked to explain the source and ownership of funds held in accounts allegedly associated with her. The open letter said materials awaiting further verification indicated that accounts linked to Li may have held assets worth tens of millions of dollars and showed records of withdrawals conducted in batches.

The statement did not present the unverified information as proof of wrongdoing and explicitly said no criminal characterization should be applied to any individual before the evidence was established.

Instead, the authors asked Li to confirm whether the accounts existed and, if so, identify who owned the assets, where the funds originated, why they entered the accounts, where withdrawals were sent and whether any of the money had a connection to BitMart customer assets.

“If these records are fake, please publicly clarify,” the statement said.

The open letter also referred to social media posts showing luxury purchases associated with Li but acknowledged that such spending was not evidence of a crime. Its authors argued that questions about the source of funds should be answered because of the allegations surrounding BitMart’s finances.

Any funds potentially connected with customer assets should be traced across accounts, companies and ownership structures, according to the statement, which called for an independent investigation into the relevant transactions.

Employees demand unpaid salaries and compensation

Employee payments form a separate part of the demands, with the statement claiming that some BitMart staff had not received their previous month’s salaries or compensation owed following the exchange’s decision to wind down.

The authors argued that ordinary employees were not responsible for decisions about company finances or the shutdown and should not bear losses resulting from management decisions.

“Work done deserves pay. Compensation owed must be paid,” the statement said.

BitMart’s closure came after the exchange had maintained a sizeable presence in crypto trading. Previous coverage in December 2025 found that BitMart showed higher order-book depth across observed Bitcoin and Ethereum perpetual markets than several competing centralized exchanges during the measured period.

The exchange also suffered a major security breach years before its current wind-down. In December 2021, hackers compromised BitMart hot wallets and removed roughly $196 million in crypto assets, after which the exchange said affected customers would be compensated.

BitMart repayment plan sought by Aug. 19

The open letter set Aug. 19 as the deadline for BitMart to provide both a verifiable asset disclosure and a detailed repayment plan for users.

Under the requested plan, BitMart would disclose the value of remaining assets and total liabilities, estimate how much customers could recover and explain the order in which repayments would be processed.

Users also requested dates for the beginning and completion of repayments, details of the party responsible for overseeing the process and confirmation of whether BitMart would submit to an independent third-party audit.

Reserve disclosures have become common among major centralized exchanges, although the scope of reporting differs between platforms. June reserve reports from Bybit and OKX showed higher customer Bitcoin balances at both exchanges, while their reported USDT holdings declined. The snapshots provided wallet and customer-balance information but did not establish the companies’ complete financial positions.

For BitMart, the open letter sought a disclosure specifically tied to its ability to meet outstanding customer claims during the wind-down, including the amount of usable reserves and liabilities remaining on the platform.

If BitMart does not provide a complete and verifiable response by Aug. 19, the authors said they would consider submitting available materials, transaction leads and other evidence to law enforcement agencies, regulators, lawyers and media organizations in multiple jurisdictions.

The statement also called on crypto companies, industry figures, researchers and media organizations to follow the dispute and support independent examination of the fund flows. Its authors said they were not asking third parties to accept the allegations in advance and instead wanted the underlying evidence made public.

“If BitMart’s core management thinks there’s any misunderstanding in the above questions, they can absolutely address them one by one with public, verifiable evidence,” the statement said.

SafePal, a Binance-backed non-custodial wallet provider, has disclosed a data breach exposing the personal details of 39,798 users, including names, shipping addresses, phone numbers, and purchase data.

According to the company’s announcement, the breach impacted customers who placed orders with SafePal between March 2, 2025, and April 11, 2026. However, SafePal has assured users that private keys, seed phrases, and crypto assets are not compromised.

Crypto Hit By Safepal Breach

SafePal disclosed the flaw on X, attributing it to a flaw in the order-tracking plug-in that exposed the personal details of a small subset of customers. According to the post, the order information of customers who placed orders between March 2, 2025, and April 11, 2026, including names, shipping addresses, email addresses, phone numbers, and purchase details, was compromised.

“Dear community, while your SafePal wallet, seed phrase, and private keys are secure, we identified a flaw in the order-tracking plug-in that led to unauthorized access to information of a subset of customers.”

The company notified the affected users via email and stated that it had identified and fixed a verification defect in the plug-in that allows customers to track their orders. SafePal has also introduced additional security measures and removed over 30 fake websites and phishing links associated with the breach. However, a report by BleepingComputer states that at least one threat actor is selling stolen data from the breach on a cybercrime forum. Additionally, some users have reported phishing attempts as early as May.

Seed Phrases, Private Keys Secure

SafePal confirmed that seed phrases, private keys, wallet passwords, hardware wallets, and crypto assets were not compromised during the breach thanks to its cold storage architecture. Furthermore, the breach did not involve bank details, payment card numbers, or any government-issued identification number. SafePal has set up a dedicated tool for users to check if their details were compromised during the breach.

Implication For Safepal Users

While the breach did not compromise users’ funds or private keys, it exposed crucial personal details tied to users. This puts users at risk of phishing attacks or elaborate social engineering scams. SafePal has warned users to be wary of attempts to access wallet credentials, crypto assets, and other personal information through fraudulent emails, text messages, phone calls, letters, offers, phishing websites, fake firmware update requests, and customer support communication.

SafePal has also issued an advisory stating it would never ask for their recovery phrase, PIN, or private keys. The advisory added that users must move their funds to a new wallet if they had entered a seed phrase or private key on a suspicious website or in response to a suspicious message following the breach.

Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

Pi Network tops crypto social dominance rankings while trading 97% below its all-time high. The gap between the project’s 60 million users and its $0.09 price reveals a structural disconnect between community attention and market demand.

Pi Network has spent much of 2026 as one of the most discussed tokens in crypto. By several on-chain social metrics, it is the most discussed. Santiment’s social dominance tracker, which measures the share of total crypto conversation that a single asset captures, has shown PI at or near the top for multiple weeks running. The project’s Telegram groups remain among the largest in crypto. Its X mentions routinely outpace tokens with ten times its market capitalization. On Reddit, Pi Network threads generate more engagement than coverage of most top-20 assets.

None of this has moved the price.

PI sits near $0.09 as of mid-August 2026, roughly 97% below the $3.00 peak it reached during the euphoria of its open mainnet launch in February 2025. The market capitalization hovers just below $1 billion. Daily trading volume rarely exceeds $15 million, a number that would be unremarkable for a token ranked in the 200s, let alone one that claims 60 million users. The gap between Pi’s social presence and its market performance is not a mystery waiting to be solved. It is a case study in the mechanics of why attention, community size, and online engagement do not automatically convert to buying pressure.

This article examines those mechanics. For readers looking for the regulatory and protocol upgrade picture, crypto.news has covered that angle separately in a feature on Protocol 27 and its implications. The focus here is narrower and, for holders, possibly more uncomfortable: what specifically breaks the transmission mechanism between social buzz and token price, and whether Pi’s version of this problem is fixable.

The social dominance paradox

Social dominance in crypto analytics refers to the percentage of total social media mentions across the industry that a single token captures. When Santiment shows Pi Network at number one, it means more people are talking about PI on Twitter, Telegram, Reddit, and other tracked platforms than about Bitcoin, Ethereum, or Solana. That is a remarkable achievement for any asset, and it is especially notable for one trading under ten cents.

The instinct is to treat social dominance as a leading indicator. If enough people are talking about a token, the logic goes, some of them will buy it, and price follows attention. This model works in specific conditions: when the token is liquid, when it trades on exchanges that the audience actually uses, and when the social activity reflects new interest from participants who do not already hold the asset. Pi meets none of these conditions cleanly.

The 60 million Pioneers who make up Pi’s user base are not passive observers discovering a new token. They are existing holders, many of whom have been mining PI on their phones for years. When they post about Pi on social media, they are not signaling fresh demand. They are expressing existing conviction. The social dominance metric captures the volume of their voices without distinguishing between a thousand new buyers researching a token and a million existing holders defending their position.

This distinction matters because social dominance correlates with price only when it reflects capital rotation. When Bitcoin’s social dominance spikes during a halving cycle, it typically coincides with new retail and institutional money entering the market. When a meme token trends on X, it often reflects a burst of speculative buying from traders who did not hold the asset before. Pi’s social activity is structurally different. It is a closed loop of community engagement that rarely intersects with the order books where price is actually determined.

Supply at zero cost: the 1.21 billion token overhang

The most direct explanation for why social buzz does not move PI’s price is mechanical: the supply side overwhelms any demand the community generates.

Pi Network’s token unlock schedule for 2026 releases roughly 1.21 billion tokens into circulation over the course of the year. That works out to approximately 6.5 million new tokens per day, every day, regardless of whether anyone is buying. An additional 775 million tokens are expected to enter the market as three-year lockup periods expire. The total circulating supply already exceeds 11 billion tokens.

The critical detail is the cost basis. Every one of these tokens was mined for free on a mobile phone. The holders paid nothing to acquire them. In any market, when a large number of participants hold an asset at zero cost, the rational behavior is to sell at any price above zero. Not all holders will sell. But enough will sell, consistently, to create a permanent headwind that requires substantial new buying to overcome.

Consider the math. At $0.09 per token, 6.5 million daily unlocks represent approximately $585,000 in potential new sell pressure every single day. That is $4 million per week, $17 million per month. For a token with daily trading volume between $10 million and $15 million, absorbing even a fraction of that sell flow requires buyers who are actively choosing to purchase PI on an exchange. Social media posts, no matter how enthusiastic, do not place buy orders.

The first year of Pi’s open mainnet proved this dynamic conclusively. Despite sustained community engagement and multiple protocol upgrades, the price fell from $3.00 to under $0.10 as unlocks flooded the market. The community grew louder as the price fell, which is exactly what you would expect from holders defending an underwater position. The social metrics improved. The price did not.

The exchange gate: why Binance and Coinbase matter more than volume

If social dominance does not convert directly to price, the natural follow-up question is: where does buying pressure actually come from? In crypto, the answer is almost always exchange listings. When a token gains access to a major exchange, it gains access to that exchange’s entire user base, millions of potential buyers who could not previously purchase the asset even if they wanted to.

Pi has made partial progress here. Kraken listed PI for spot trading in March 2026, and OKX extended access to U.S. users in May. Both were meaningful milestones. But as crypto.news detailed in its price prediction analysis, the token kept falling after both listings. The reason is that Kraken and OKX, while reputable, are not where most retail crypto buyers live.

Binance and Coinbase together account for a disproportionate share of global retail trading volume. Binance’s user base exceeds 200 million registered accounts. Coinbase serves as the default entry point for American retail investors. A token listed on both platforms has access to a liquidity pool that is qualitatively different from one available only on mid-tier exchanges.

Binance held a community vote in February 2025 in which 86.8% of roughly 226,000 participants supported listing PI. The exchange never acted on the result. The stated concerns, code transparency, insufficient independent security audits, questions about decentralization, and token concentration risk, remain unresolved as of August 2026. Coinbase has been even quieter, offering no public commentary on PI at all.

The absence of these two platforms creates a structural ceiling on demand. Pi’s community can generate all the social buzz in the world, but if the exchanges where most buyers transact do not offer PI, that buzz has no on-ramp to the order book. The community is loud. The order book is thin. And price is set by the order book.

How social buzz without liquidity traps price

The interaction between high social activity and low liquidity creates a specific kind of trap. In a liquid market, social attention leads to order flow, which leads to price discovery. In an illiquid market, social attention leads to frustration, which leads to more social activity, which still does not generate order flow. The feedback loop amplifies noise without amplifying signal.

Pi’s trading volume tells this story clearly. Daily volume sits between $10 million and $15 million, with occasional spikes above $20 million during catalyst events. For context, Dogecoin, a meme token with a fraction of Pi’s claimed user base, routinely trades $500 million to $1 billion per day. Shiba Inu, another community-driven token, regularly sees $200 million or more. The difference is not community size. It is exchange access and speculative capital flow.

When Pi spikes on a catalyst, the pattern is consistent. Protocol v25 triggered a 39% rally in July 2026, pushing PI briefly above $0.10. Within days, the rally lost steam as open interest collapsed to $9.6 million and sellers absorbed the move. The CPI data release on August 12 pushed PI up 5% in a single session, then the token drifted back toward $0.088. Each spike attracts social media celebration, which registers as rising social dominance, which commentators interpret as bullish, which does not produce a sustained bid.

The thin order books mean that even modest selling, a few hundred thousand dollars, can push the price down noticeably. Conversely, thin books also mean that a genuine demand shock, a Binance listing, for example, could move price dramatically upward. But that is a statement about potential, not current reality. In the current structure, PI is caught in a low-liquidity trap where social engagement circulates within the community without reaching the exchange infrastructure that determines price.

Historical precedent: large communities, thin markets

Pi is not the first token to have a massive community and a declining price. The history of crypto is littered with projects that built enormous social followings only to watch price detach from engagement. The patterns are instructive.

Internet Computer (ICP) launched in May 2021 at a peak near $700, backed by a sophisticated technical vision and a community of developers who had followed the project for years. Within three months it had fallen to $30. Within two years it was under $5. Throughout that decline, the ICP community remained one of the most vocal in crypto, producing constant content about the project’s technical merits. Social activity stayed high. Price kept falling. The mechanism was the same as Pi’s: massive token unlocks from early participants who had received allocations at low or zero cost, combined with a market that had already priced in the best-case scenario before the fundamentals could catch up.

EOS raised $4 billion in the longest ICO in crypto history, launched with one of the largest and most engaged communities in the industry, and spent the next five years losing more than 95% of its value. The EOS community produced more governance proposals, more social media content, and more developer advocacy than most projects in the top 100. None of it translated to sustained buying pressure because the token’s supply dynamics and competitive position deteriorated faster than community enthusiasm could compensate.

XRP presents a different but related lesson. Ripple’s community, often called the XRP Army, has been one of the most active social media forces in crypto for nearly a decade. During the SEC lawsuit years from 2020 to 2025, XRP’s social dominance routinely spiked to levels comparable to Bitcoin and Ethereum. Price moved only when external catalysts, court rulings, exchange re-listings, provided structural changes to accessibility and regulatory risk. The social activity itself was noise. The signal was the legal and exchange infrastructure underneath it.

The common thread across all three cases is that community engagement sustains attention but does not create the structural conditions for price appreciation. Those conditions require some combination of reduced supply growth, expanded exchange access, and genuine on-chain utility. Pi has the attention. It is still working on the rest.

The opposing case: when community mass did convert

Not every large community fails to move price. Dogecoin and Shiba Inu both started as community-driven projects with minimal technical differentiation and achieved market capitalizations in the tens of billions. Understanding what made them different from Pi clarifies what Pi would need to change.

Dogecoin’s 2021 rally was driven by a specific set of conditions that Pi does not share. First, DOGE was listed on every major exchange, including Binance, Coinbase, and Robinhood, giving its community direct access to the deepest liquidity pools in crypto. Second, the community’s social activity attracted new capital from outside the existing holder base, driven in part by endorsements from Elon Musk and viral TikTok campaigns. Third, DOGE’s supply inflation, while perpetual, was low relative to its market cap, meaning the dilution did not overwhelm incoming demand.

Shiba Inu followed a similar pattern. Its community generated enough social momentum to force exchange listings, which created the liquidity infrastructure that allowed social buzz to convert to order flow. The token was listed on Binance within months of its initial surge and on Coinbase shortly after. Each listing unlocked a new pool of retail buyers.

The critical difference is the sequence. For DOGE and SHIB, community energy led to exchange listings, which led to liquidity, which led to price appreciation. For Pi, the sequence is stalled at the second step. The community energy exists. The exchange listings, at least on Binance and Coinbase, have not followed. Without the liquidity bridge, the community’s energy circulates internally without converting to market demand.

The opposing thesis for Pi bulls is straightforward: if Binance or Coinbase lists PI, the dynamic could reverse rapidly. Pi’s community is larger than Dogecoin’s was at the time of its 2021 breakout. If that community gains access to deep order books, the pent-up demand could produce a price move that dwarfs anything Pi has seen since its mainnet launch. This thesis is invalidated if both Binance and Coinbase continue to decline PI after Protocol 27 stabilizes the protocol and the ESMA registration removes the EU regulatory question, because at that point the community will have run out of structural excuses. It is also invalidated if a major listing occurs and the price still falls, which would confirm that the supply overhang is simply too large for any amount of retail demand to absorb.

What Protocol 27 and the ESMA filing change, and what they do not

Two recent developments have renewed optimism within the Pi community: the approaching Protocol 27, which the Core Team has called the “final planned upgrade,” and the ESMA registration of Pi’s MiCA white paper (entry 549, filed by PiBit Ltd). Both are real milestones. Neither directly addresses the social-to-price disconnect.

Protocol 27 matters for exchange listings because a stable, finalized protocol is easier to audit than one undergoing frequent breaking changes. Exchanges cite code transparency and security audit concerns as reasons for not listing PI. A protocol that stops changing gives independent auditors a fixed target, which could eventually produce the third-party security review that Binance and Coinbase appear to require. But Protocol 27 itself is not an audit. It is a precondition for one.

The ESMA registration matters because it provides legal standing for PI to be offered within the European Union. For exchanges considering EU markets, this removes a regulatory blocker. But dozens of tokens have registered MiCA white papers. The registration makes Pi compliant with a baseline requirement. It does not differentiate the project from competitors.

Neither development changes the supply schedule. The 6.5 million daily token unlocks will continue after Protocol 27 and after ESMA registration. Neither development forces Binance or Coinbase to list PI. And neither development converts social media engagement into exchange order flow.

What these milestones do is narrow the list of valid objections. Before Protocol 27, critics could argue that the protocol was too immature for serious exchange integration. After Protocol 27, that argument weakens. Before the ESMA filing, critics could argue that Pi lacked regulatory standing in major markets. After the filing, that argument is gone for the EU. The community’s task shifts from generating noise to forcing a decision from the exchanges that control access to retail liquidity. Whether that decision comes in Pi’s favor depends on factors, independent audits, governance reform, token concentration analysis, that social dominance cannot influence.

What to watch

Four metrics will determine whether Pi’s social dominance eventually converts to demand or remains permanently decorative.

First, watch the Binance and Coinbase response to Protocol 27. If the protocol stabilizes and both exchanges still decline to list PI within six months, the structural barrier to demand conversion is likely permanent under current conditions. Every month without a listing is a month where 195 million new tokens enter circulation without a matching increase in buyer access.

Second, watch daily trading volume relative to daily unlocks. If volume consistently stays below $20 million while 6.5 million tokens per day enter circulation, the sell-side math remains unfavorable. A sustained rise above $30 million to $50 million in daily volume, even without a major listing, would suggest that organic demand is beginning to absorb supply.

Third, watch on-chain transaction activity distinct from exchange trading. Pi’s claimed 13 million active wallet addresses and 51,800 Pioneer-built applications represent potential utility. If those applications generate real transaction volume, measured by contract calls, not just wallet counts, Pi would have a demand source independent of exchange listings. The KYC validator workforce that completed 526 million tasks is an example of real on-chain utility, but its scale, roughly $2 million in total payouts, is too small to move a billion-dollar token’s price.

Fourth, watch the composition of social activity. If Pi’s social dominance begins to include mentions from institutional accounts, exchange research desks, and DeFi protocols, that signals a broadening of interest beyond the existing holder base. If social dominance remains driven entirely by Pioneers defending their position, the metric is measuring conviction, not demand. Conviction without liquidity is a community. Conviction with liquidity is a market.

Volume, meanwhile, is dead simple as it measures the number of contracts that changed hands during a given period. Think of it as measuring how many times the front door of that exclusive club opened and closed over a given period, regardless of who stayed. It thus represents the degree of churn or liquidity available to manage positions.

So, the latest case of volume falling far behind OI is like a large club with a tiny exit door. What happens if a large number of people try to rush out?

Because overall investor positioning is massive, a sudden catalyst could trigger a wave of contract closures, such as forced liquidations due to margin shortages. Without the underlying daily volume to provide liquidity, the market may not be able to absorb the rush smoothly, leading to volatile, exaggerated price swings.

“The risk is mechanical. When open interest towers over daily volume, liquidations meet little resting flow to absorb them, and adverse moves extend further than they otherwise would. Traders have added substantial risk, most of it long, into a market that shows no matching demand,” blockchain analytics firm Glassnode said in a report.

The risk of an exaggerated move is particularly likely to the downside because of weakening demand and a lack of resting bids or buy orders at lower price levels.

Binance has overtaken CME Group in Bitcoin futures open interest for the first time since late 2023, holding roughly 148,500 BTC against CME’s 102,840. The reversal unwinds two years of institutional dominance narrative and raises questions about whether traditional finance is retreating from crypto derivatives or simply relocating.

For two years, a single chart told the story of Bitcoin’s institutional coming of age. CME Group, the Chicago exchange where pension funds, sovereign wealth managers, and hedge funds trade everything from corn to crude oil, held more Bitcoin futures open interest than any venue on Earth. That lead over Binance, the offshore exchange synonymous with retail speculation, became the most cited proof point for the “institutions are here” thesis.

That chart has now flipped. Binance holds roughly 148,500 BTC in open interest, worth approximately $9.6 billion. CME has dropped to around 102,840 BTC, or $6.7 billion, its lowest reading since February 2024. The gap is not narrow. It is roughly 45,000 BTC wide and growing.

The shift did not arrive overnight. CME open interest has fallen for five consecutive months, accelerating through the second quarter of 2026 as the profitability of the basis trade collapsed and institutional appetites shifted. What looked like a permanent structural change in Bitcoin market microstructure may have been, at least in part, an arbitrage play dressed in institutional clothing.

Understanding what happened, why it matters, and where it leads requires following the money through a maze of basis spreads, regulatory upheaval, and the evolving definition of what “institutional” even means in crypto.

The basis trade machine and how it broke

The centerpiece of CME’s rise to the top of the Bitcoin futures leaderboard was not directional conviction. It was the cash and carry basis trade, a delta neutral strategy older than most of the people trading it.

The mechanics are straightforward. Buy spot Bitcoin, or more commonly after January 2024, buy shares of a spot Bitcoin ETF like BlackRock’s IBIT. Simultaneously sell Bitcoin futures on CME at a premium to the spot price. The difference between the futures price and the spot price, the basis, represents annualized yield. When Bitcoin was rallying through 2024 and the first half of 2025, that basis regularly exceeded 15% to 20%, dwarfing anything available in traditional fixed income.

Hedge funds, proprietary trading desks, and institutional players rotated capital into this trade at scale. According to CFTC Commitments of Traders data, leveraged funds held persistent net short positions on CME Bitcoin futures throughout most of 2024 and 2025, the signature footprint of the basis trade. They were not bearish on Bitcoin. They were harvesting yield from the contango.