In June 2026, four mining pools controlled 74% of Bitcoin's total hashrate. Foundry USA led with 31%, AntPool with 18%, ViaBTC with 13%, and F2Pool with 10%—data from miningpoolstats.stream, a source I have used since my first whitepaper audit in 2017. The network designed to distribute trust has become a study in concentrated power.
This is not a sudden coup. It is the inevitable outcome of the 2024 halving, the subsequent profit squeeze, and the quiet evolution of mining from a garage hobby to a institutional balance sheet asset. The narrative of 'mining is democratic' has long been a fond memory. What we have now is a two-tier system: one for the whales, one for the rest.
The Institutional Layer: Service, Not Just Hashrate
Foundry USA, backed by Digital Currency Group, operates with strict KYC and compliance—a feature, not a bug, when courting pension funds and ETF issuers. Its 31% share corresponds to roughly 2.62 EH/s, and it does not publish its fee structure for large clients. Based on my experience coordinating dialogues between BlackRock and DAOs in 2025, I can confirm that institutional miners pay a premium for tax reporting, transaction filtering, and priority settlement. The advertised 4% PPLNS fee is for retail; the real rate for a 50 MW facility is negotiated behind closed doors.
AntPool, owned by Bitmain, leverages hardware lock-in. Buy an Antminer S21 Pro, and the firmware nudges you toward its pool. ViaBTC has faced regulatory scrutiny in Europe, forcing it to upgrade KYC—a cost that smaller pools cannot easily absorb. F2Pool, the oldest, maintains global low-latency servers but has seen its share slip as institutions gravitate to compliant partners.
The result: the top four pools offer a tier of service that includes dedicated account managers, customized payout schedules, and even balance-sheet financing for equipment. Small miners—those with a few hundred machines—get the same 4% fee and a dashboard.
The Retail Layer: Fee Sensitivity and the EMCD Experiment
Enter EMCD, a pool claiming nine years of experience and a fee as low as 1.5%—a 62.5% discount versus the industry standard. It promises equal treatment for miners of all sizes. As of June 2026, it held 2.7% of global hashrate, roughly 0.23 EH/s.
Trust no one. Verify everything.
Is 1.5% sustainable? A typical pool’s operating costs include server infrastructure, DDoS protection, payment processing, and compliance. For a small pool, these can consume 1–2% of revenue. EMCD must either operate on razor-thin margins or rely on ancillary income—maybe bundled hosting, maybe future tokenization. I have seen this playbook before. In 2020, a similar low-fee pool in Asia collapsed after three months, leaving miners unpaid. The difference is that EMCD claims nine years of industry presence, which suggests a survival instinct that pure newcomers lack.
The contrarian angle is uncomfortable: perhaps centralization is a feature, not a bug. Institutional pools reduce counterparty risk for large capital. They offer transaction transparency and regulatory compliance that allow Bitcoin to integrate with traditional finance. The 74% concentration may be the price of mainstream adoption.
But that price is paid by the small miner. If the network’s security is increasingly controlled by a handful of entities subject to US sanctions policy—Foundry has already filtered transactions from OFAC-listed addresses—then Bitcoin’s censorship resistance becomes conditional. Gold is heavy. Code is light. But code bound by law is just a heavier ledger.
Summer fades. Builders remain.
The Structural Path Forward
The next six months will determine two things: whether EMCD can scale its low-fee model without breaking, and whether the top four pools will respond with more segmented offerings. I suspect the latter. Foundry may launch a 'mini-institutional' tier for medium-sized miners, with a 3% fee and a lighter KYC process. AntPool could bundle its next-gen miners with a two-year pool loyalty discount.
Alternatively, if EMCD’s share rises to 5–7% and maintains clean payment records, it will validate the thesis that a non-aligned, low-cost pool can disrupt the oligopoly. That would be a signal worth watching. Noise is cheap. Signal is rare.
I wrote in 2017 that 'Math Over Hype' would define the space. Now, the math is clear: 74% of the network’s security sits in four baskets. The question is not whether the baskets are secure, but whether the network can afford to let them break.
Faith requires reason. Reason requires data. The data says the next move belongs to the miners—and to the pools that serve them, not the other way around.
