Binance Pool Shuts Down: Bitcoin Mining Pools in 2026

On October 13, Binance told its miners the run was over. In a short notice, the world’s largest exchange said it would shut Binance Pool and its related mining products at 07:00 UTC on November 10, 2026, ending a business it had launched back in April 2020. As Decrypt reported, the closure sweeps in the pool itself plus Cloud Mining, Pool Savings and the Transaction Accelerator. Six years after exchanges raced to plant flags in Bitcoin mining, the last of the big exchange-run pools is quietly walking away.

The hashrate at stake is small. Binance Pool points somewhere between 6 and 14 EH/s at the network, barely over 1 percent of Bitcoin’s roughly 1,085 EH/s of total computing power, according to the Hashrate Index pool tracker. On its own, the closure will not reshape the leaderboard. What makes it worth a close look is the pattern around it. SBI Crypto, the mining arm of Japan’s SBI Group, shut its public pool at the end of July. CoinDesk covered that exit too. Poolin, once a top-three pool, froze customer withdrawals in 2022 and never truly recovered, as Bitcoinist documented at the time. The middle of the pool market is thinning, and hashrate keeps flowing toward a handful of large operators at the top.

This piece uses the Binance exit as a way into the whole subject. What is a mining pool, really? How does it pay the people who point machines at it? Who runs the biggest ones, why does their concentration worry so many Bitcoiners, and what does the current wave of closures say about where mining is heading? Figures here are current as of October 28, 2026, with Bitcoin trading near $80,600, about 36 percent below its October 2025 record of $126,080, per CoinGecko.

What Binance Pool’s shutdown actually means

The notice itself is brief and bureaucratic. Binance said only that “as part of our ongoing review of products and services” it would discontinue Binance Pool and related services on November 10. No commercial, financial or regulatory reason was offered. FinanceFeeds noted that the company gave miners nothing beyond that boilerplate.

The shutdown is wider than the pool alone. At the cutoff, Binance switches off the pool, Cloud Mining, Pool Savings and the Transaction Accelerator, along with the supporting plumbing: the Stratum connection URLs, the mining API, and the email and Telegram hashrate alerts, according to Tech Times. Miners who do nothing will simply be disconnected the moment those endpoints go dark.

For anyone still mining on Binance Pool, the checklist is short but time-boxed:

• Repoint hardware to another pool before 07:00 UTC on November 10 to avoid a gap in rewards.
• Settle outstanding balances, pending rewards, refunds and any required identity verification by 23:59 UTC on November 10.
• Export anything you need from your account; mining records stay visible for 14 days, until November 24.

Then there is the matter of scale. At the end of 2025, Binance Pool was the ninth-largest pool with about 26 EH/s, roughly 2.3 percent of the network. By late October 2026 it had slipped to around 13 to 14 EH/s, close to 1.3 percent. It was never a giant. But it was a familiar name with a low barrier to entry, and its departure removes one more independent venue for small miners to send work. As the mining-data firm Hashrate Index put it in analysis cited by Decrypt, Binance Pool was “the last pool still standing from the era when exchanges rushed into mining.”

What a mining pool is, and why solo miners need one

Strip away the jargon and Bitcoin mining is a lottery that draws a winner about every 10 minutes. Specialized machines called ASICs race to find a number that produces a valid block hash; the first to succeed wins the right to add the next block and collect its reward, currently 3.125 BTC (about $252,000 at today’s price) plus the fees on the transactions packed inside it. The network retargets its difficulty every 2,016 blocks so that, no matter how much computing power joins, the draw keeps landing near that 10-minute mark. In mid-October the difficulty rose about 6.4 percent to roughly 132.7 trillion, a record, as hashrate pushed past 1 zettahash for the first time, Pickaxe reported.

Here is the problem for an individual. The odds of any single machine winning are proportional to its share of total hashrate, and total hashrate is now astronomical. A current-generation ASIC running at about 200 TH/s controls roughly one part in 5.4 million of the network. On average it would expect to find a block about once a century. A hobbyist’s Bitaxe at around 1 TH/s faces odds closer to one in 20,000 years. The expected value is fine (over a long enough horizon the arithmetic works out), but the variance is brutal: you could mine for decades and earn nothing at all.

A mining pool solves that variance problem by turning a lottery into a salary. Thousands of miners point their machines at one pool, which combines their hashrate into a single enormous entry. When any member finds a block, the pool collects the reward and distributes it to everyone in proportion to the work they contributed, measured in “shares,” the near-miss hashes that prove a miner was genuinely trying. Instead of waiting a lifetime for a jackpot, each participant gets small, steady, predictable payouts. The pool keeps a cut, usually between zero and 4 percent, for running the service and carrying the risk.

That last point reframes the Binance news. A pool is less a piece of mining hardware than a financial product: its core job is to smooth out variance, which is another way of saying it sells insurance against bad luck. Understanding pools as risk businesses, not server farms, explains almost everything that follows, from how they price their fees to why an exchange might decide the whole thing is no longer worth the trouble.

How miners get paid: PPS, FPPS, PPLNS and the rest

Payout schemes differ mainly along one axis: who absorbs the variance, the pool or the miner. Under Pay-Per-Share (PPS), the pool pays a fixed amount for every valid share whether or not a block is found, so the operator eats all the luck and charges the highest fee for the privilege. Full-Pay-Per-Share (FPPS) works the same way but folds an average of transaction fees into the payout as well; it has become the default for large institutional farms that want bankable, predictable income. PPS+ splits the difference, paying the subsidy PPS-style while distributing transaction fees only when the pool actually mines a block.

PPLNS (Pay-Per-Last-N-Shares) flips the risk back onto the miner: you are paid only when the pool finds a block, based on your recent shares, which makes income lumpier but fees lower and discourages “pool hopping.” Newer designs go further. Ocean’s TIDES pays transparently and non-custodially straight from the coinbase transaction, while SOLO options route an entire block to whoever gets lucky, minus a small cut. The table below summarizes the main choices; per-pool fee data is drawn from D-Central’s 2026 pool comparison.

Fees track the risk the operator takes on. Foundry charges large miners effectively nothing on FPPS, using the pool as a loss-leader for a wider business; AntPool and F2Pool sit near 2.5 percent; ViaBTC runs 4 percent on PPS+ and 2 percent on PPLNS. The gap between a pool’s headline fee and what a miner actually nets is its own small study, not unlike what a perp DEX trade really costs once you add up the pieces nobody advertises.

Who runs Bitcoin’s pools in late 2026

A pool’s name on the leaderboard is a logo, not necessarily an independent company. Here is the live picture as of October 28, 2026, drawn from the Hashrate Index pool tracker against a network running near 1,085 EH/s, with Bitcoin around $80,750 at the time of the reading.

The ownership map matters more than the ranking. Foundry belongs to Digital Currency Group, the US conglomerate behind Grayscale and others. AntPool is run by Bitmain, the company that also manufactures most of the world’s ASICs, so the same firm sells the shovels and runs one of the biggest claims. ViaBTC was founded by Haipo Yang, who the following year launched the exchange CoinEx; pool and exchange remain two arms of one business. MARA Pool is the in-house pool of a Nasdaq-listed miner, and Ocean is the non-custodial outsider seeded by Block founder Jack Dorsey. Count operators rather than logos and the market looks tighter still, a point we will return to.

The exchange-pool era is ending

When Binance Pool launched in April 2020, exchanges were rushing into mining. The logic was simple: an exchange already held customer coins, ran custody, and had order books hungry for supply, so a pool looked like a natural funnel. Freshly mined coins could flow straight into trading and lending. Huobi, OKEx and others built pools on the same reasoning. For a few years it worked.

It stopped working for several reasons at once. Mining margins compressed as difficulty kept climbing and the 2024 halving cut the block subsidy to 3.125 BTC. A soft market left Bitcoin about 36 percent below its 2025 peak, squeezing every operator. Running a competitive pool now demands genuine engineering, from FPPS variance-underwriting to low-latency Stratum infrastructure to institutional onboarding, none of which plays to an exchange whose profit comes from trading fees. And regulatory scrutiny turned exchange-linked mining from an asset into a liability. Seen that way, Binance Pool is not an outlier; it is the latest name in a steady procession.

The through-line, as Hashrate Index framed it, is that the exchange era in mining is drawing to a close. Binance is the clearest case. The company is pivoting hard toward traditional finance: it has rolled out thousands of tokenized US equities and ETFs through its bStocks product and shut its NFT marketplace earlier in 2026, per Decrypt. Against that backdrop, a low-margin pool weighed down by hardware support and compliance overhead simply no longer fits the strategy.

Where does Binance Pool’s hashrate go?

Those 13 to 14 EH/s have to land somewhere before November 10. Displaced hashrate tends to follow the path of least resistance, which means the biggest, most reliable pools with the lowest fees and the smoothest onboarding. When SBI Crypto closed, it pointed its miners toward Braiins and Luxor. Binance’s own guidance, as relayed by Tech Times, named Foundry, AntPool, ViaBTC, F2Pool, SpiderPool and Ocean as alternatives.

The catch is that “biggest, most reliable, lowest-fee pool” describes Foundry and AntPool better than anyone else, so each closure nudges the network a little more concentrated unless miners deliberately choose smaller or non-custodial homes. A miner leaving Binance Pool can absolutely pick Ocean, which is built around decentralization, or a mid-tier independent, but the gravitational pull runs toward the top of the table. That is the quiet cost of consolidation: not a dramatic takeover, just hashrate drifting upward, pool exit by pool exit.

One silver lining deserves stating plainly. An orderly, pre-announced shutdown, with weeks of notice and a clear path to withdraw balances, is the good kind of pool exit. The bad kind looked like Poolin in 2022, which had quietly been yield-farming customer float and froze withdrawals when the market turned against it. Binance’s miners get time to plan their move. Poolin’s did not.

Concentration, the Nakamoto coefficient and the 51% question

The leaderboard explains why pool exits make Bitcoiners uneasy. Add the top two pools, Foundry and AntPool, and you reach roughly 47 percent of the network. Add the third, F2Pool, and you cross 63 percent. The top four clear 70 percent, and about six pools together mine more than 80 percent of all blocks, a concentration News.Bitcoin.com has tracked through the year.

The standard shorthand for this is the Nakamoto coefficient: the smallest number of entities that would have to collude to control a majority of hashrate. For Bitcoin’s pools in late 2026, that number is three. Three operators, acting together, could in principle command the majority of the network’s block space. The pattern is not unique to mining; concentration-of-control questions shadow every corner of crypto, from gaming DAOs and governance capture to who gets to issue the biggest stablecoins.

This is where the dreaded “51% attack” enters. A party controlling a majority of hashrate could, as Chainlink explains, rewrite recent blocks, double-spend coins by reorganizing the chain, and censor or delay transactions. What it could not do is steal coins from wallets it does not control or mint Bitcoin beyond the fixed schedule; the cryptography protecting balances is untouched. The threat is to the immutability of recent history, not to the ownership of individual accounts.

Two things keep the horror scenario mostly theoretical. The first is economics. Fresh issuance alone runs about 450 BTC a day (3.125 BTC across 144 blocks), worth roughly $36 million at current prices, and an attacker would be setting fire to the value of the very asset and hardware it depends on. The second is that pools are not monolithic. The hashrate belongs to thousands of individual miners who can repoint their machines within minutes if an operator misbehaves. GHash.io briefly touched 51 percent in 2014 and voluntarily capped itself after the backlash; that social pressure remains Bitcoin’s first line of defense.

Concentration may also understate itself. Research by the Bitcoin developer known as 0xB10C has shown that several nominally separate pools broadcast near-identical block templates, suggesting some may be the same operator under different logos, a finding laid out in his template-similarity analysis. If AntPool and its apparent satellites are counted as one entity, the effective Nakamoto coefficient may sit closer to two than three. The leaderboard, in other words, probably overstates how many independent hands actually build Bitcoin’s blocks.

Who really decides what goes in a block

Here is the subtlety that makes pool concentration more than an abstract worry. Under the long-standing Stratum V1 protocol, the pool, not the miner, builds the block template. The operator decides which transactions go in, in what order, and therefore which get confirmed and which keep waiting. The miner simply supplies raw hashing power against a template handed down from above.

That design made sense in 2012, when pools were small and trust was casual. In 2026, with a handful of operators directing most of the hashrate, it means a very small number of companies effectively curate the contents of nearly every Bitcoin block. Block space is valuable and getting more so. Transaction fees, the ordering of trades, and the booming world of Bitcoin-native tokens all turn on who assembles the block. The scramble for scarce space is real enough to drive whole markets, as our look at Bitcoin Runes and the race for short names showed: when inscriptions and token issuance fight for room, the entity choosing what goes in holds genuine economic power.

And whoever builds the template can also, in principle, leave transactions out. That is not hypothetical, as the next section shows.

Stratum V2, Job Declaration and Ocean’s DATUM

The fix for pool-controlled block-building is a protocol upgrade called Stratum V2. Where V1 sends plaintext jobs and lets the pool dictate the template, V2 is a binary, encrypted rewrite (roughly 70 percent smaller on the wire, with a Noise handshake and Schnorr authentication) whose headline feature is Job Declaration. The miner, running its own full node, builds the block template locally, and the pool merely validates the coinbase and smooths payouts. Block-building power moves back to the people actually doing the hashing, a shift Spark’s research describes as the most consequential change to mining’s plumbing in a decade.

Momentum is real. On May 7, 2026, seven pools representing nearly 75 percent of hashrate joined the Stratum V2 working group, Foundry, AntPool and F2Pool among them. AntPool CEO Andy Zhou said the company was “proud to support the broader adoption of Stratum V2” as an open standard for efficiency, security and decentralization, per crypto.news. The first production Job Declaration block followed on June 25, 2026, block 955,318, mined by the V2-native pool DMND with hashrate supplied by GoMining.

The people behind that block framed the stakes clearly. DMND CEO Alejandro De La Torre called it the moment “a miner just mined the first Stratum V2 block to power their own product end to end,” while GoMining CEO Mark Zalan was blunt about what it displaced: “For years, mining pools have determined which transactions are included in Bitcoin blocks,” he told TFTC. The point of Job Declaration is to end exactly that arrangement.

There is a deployment catch. Job Declaration needs new firmware that most ASICs do not yet run. Ocean’s DATUM takes a more pragmatic route, letting a miner build templates locally on top of existing V1 firmware by running a node and a gateway, with no hardware upgrade required. It is less elegant but deployable today, which is why Ocean, though small at under 2 percent of hashrate, matters to this story out of all proportion to its size. If the Binance exit pushes even a slice of displaced hashrate toward Ocean or a V2 pool rather than the top two, it would, ironically, leave block-building a little more decentralized than before.

The censorship flashpoint

Why does template control matter so much in practice? Because a pool that builds the block can be pressured to filter it. The clearest case is F2Pool. In late 2023 the developer 0xB10C noticed F2Pool was omitting transactions that touched addresses on the US Treasury’s OFAC sanctions list. F2Pool co-founder Chun Wang confirmed it without apology, asking why anyone should be surprised that he would refuse to confirm transactions for “criminals, dictators and terrorists,” in remarks documented by The Miner Mag.

After a backlash, Wang disabled the filter and, tellingly, reframed the issue as a design flaw rather than a moral stance: a censorship-resistant system, he argued, must resist censorship at the protocol level rather than relying on each participant to behave conscientiously. That is precisely the gap Stratum V2 and DATUM aim to close. If the miner builds the template, no single pool operator sits at a chokepoint where a government order, or a change of heart, can quietly drop transactions.

The censorship worry is not that Bitcoin will suddenly start blocking payments. It is structural: the more concentrated block-building becomes, the fewer places pressure has to be applied to have a network-wide effect. Every pool that closes, Binance Pool included, removes one more independent template-builder from the board, which is why decentralization advocates watch the exits with genuinely mixed feelings.

Counterparty risk: when the pool holds your coins

Most pools are custodial. They collect the block reward and hold a miner’s accrued earnings until a payout threshold is reached, which can leave meaningful balances sitting on the pool’s books. That float introduces a risk with nothing to do with Bitcoin’s cryptography and everything to do with trusting a company. Poolin is the cautionary tale: in September 2022 it froze BTC and ETH withdrawals after, by its own account, deploying customer balances into DeFi yield strategies that soured. Miners simply could not get their coins out, as Bitcoinist reported.

This is the lens through which to judge the Binance shutdown, and by that measure it mostly passes. Binance is giving weeks of notice, a hard date, and explicit instructions to withdraw balances and finish verification by November 10. An exchange with deep liquidity winding a product down in the open is about as benign as a pool closure gets. The real danger in this market is not the announced exits; it is the unannounced ones, where a stressed operator sits on customer float and prays the market turns before anyone asks for their money.

Non-custodial designs sidestep the problem entirely. Ocean’s TIDES scheme pays miners directly from the coinbase transaction of each block, so there is no pool-held balance to freeze in the first place. That is a structural answer to a structural risk, and it is part of why a vocal slice of miners will trade convenience to avoid handing custody to anyone at all.

How US regulators see mining pools

For American miners, the regulatory picture turned clearer in 2026, and it is mostly good news. In March 2025 the SEC’s Division of Corporation Finance stated that proof-of-work mining, including participation in a mining pool, is not a securities transaction: rewards come from a participant’s own computational work, an administrative or ministerial act, rather than from the entrepreneurial efforts of a third party, so it fails the Howey test, as The Block reported. A broader interpretive release in March 2026 extended that no-securities view to mining, staking and related activities. SEC Chair Paul Atkins framed the approach in plain terms in the agency’s announcement: “This is what regulatory agencies are supposed to do: draw clear lines in clear terms.”

That does not make pools unregulated. Pools that hold customer funds, sell cloud-mining products, or route rewards through an exchange start to look like money-services or custody businesses, which is where oversight bites. Binance’s wind-down of a pool bundled with savings and cloud mining, in a year when the company has faced pressure on several regulatory fronts, fits a pattern of exchanges shedding anything that adds compliance surface without adding much margin. Crypto derivatives tied to mining, such as hashrate or hashprice futures, sit with the CFTC rather than the SEC, and sanctions pressure of the kind F2Pool met comes from the Treasury, not the securities regulator. The bottom line: mining itself is clearly not a security in the US, but the business wrapped around a pool can still draw a regulator’s eye.

What to watch after November 10

Binance Pool’s lights go out at 07:00 UTC on November 10, and in the immediate aftermath the numbers to watch are straightforward. Does the combined Foundry and AntPool share tick up as displaced hashrate migrates, or do miners spread out? Does Ocean or a V2 pool pick up a visible bump? A few weeks from now the leaderboard will tell us whether this exit concentrated the network further or, against the prevailing trend, nudged a little hashrate toward the decentralizing tools now on the table.

The larger questions outlast any single closure. Will Stratum V2’s Job Declaration move from a handful of demonstration blocks to a meaningful share of real ones? Will the “AntPool and friends” clustering that 0xB10C flagged ever be confirmed or dispelled? And will the slow exit of exchange-run pools leave mining healthier, with fewer custodial middlemen and more purpose-built operators, or simply more concentrated at the very top?

For now the practical takeaway is narrow and clear. If you mine on Binance Pool, move your machines and withdraw your balance before November 10. And if you care about where Bitcoin’s block-building power sits, treat your choice of pool as a vote. In a network where three operators can already cross a majority, the pool a miner points at is not just a payout preference; it is a small, continuous decision about how decentralized Bitcoin really is.

Frequently Asked Questions

Why is Binance shutting down its mining pool?

Binance said only that it was discontinuing Binance Pool as part of an ongoing review of its products and services, with no commercial or regulatory reason given. Analysts link the move to compressed mining margins, a soft 2026 market, and Binance’s pivot toward traditional finance products like tokenized equities. It is also the last of the big exchange-run pools, a model that has steadily fallen out of favor.

When does Binance Pool close and what should miners do?

Binance Pool, along with Cloud Mining and Pool Savings, closes at 07:00 UTC on November 10, 2026. Miners should repoint their hardware to another pool before that time to avoid losing rewards, and settle any outstanding balances, refunds and identity verification by 23:59 UTC the same day. Mining records remain viewable for 14 days, until November 24.

What is a Bitcoin mining pool and how does it work?

A mining pool combines the hashrate of many miners into one large entry so they can win blocks more often and share the reward in proportion to the work each contributed. It turns mining from a rare, high-variance lottery into small, steady payouts, usually in exchange for a fee of zero to 4 percent. A single machine might otherwise wait decades or longer to find a block on its own.

Which Bitcoin mining pool is the largest in 2026?

Foundry USA is the largest Bitcoin mining pool in late 2026, at about 26 percent of network hashrate, followed by AntPool near 21 percent and F2Pool near 17 percent. Those top three together control more than half of all block production, which is why the Nakamoto coefficient for Bitcoin mining is three. The top four pools exceed 70 percent combined.

Can a mining pool launch a 51% attack on Bitcoin?

In theory a group controlling a majority of hashrate could reorganize recent blocks, double-spend, or censor transactions, but it could not steal coins from existing wallets or create new Bitcoin beyond the schedule. In practice the economics are punishing, because an attacker would destroy the value of the asset and hardware it depends on, and miners can leave a misbehaving pool within minutes. No Bitcoin pool has sustained a majority since GHash.io voluntarily capped itself in 2014.

By Yuki Tanaka, HOGE Wire mining and staking desk.