Predictability is a myth; only volatility is real. On June 30, 2026, miningpoolstats.stream published a snapshot that should have made every Bitcoin alarm sound: the top four mining pools controlled over 70% of the network’s total hashrate. Foundry USA alone commanded 31% — nearly a third of all computational power securing the ledger. AntPool held 18%, ViaBTC 13%, and F2Pool 10%. The remaining 28% was scattered among dozens of smaller pools, with EMCD, a relative newcomer, barely registering at 2.7%.
But the story isn’t just about numbers. It’s about what those numbers mean for the participants who actually run the machines — the miners themselves. Behind the aggregate percentages lies a structural rift that has been quietly widening since the 2024 halving. The mining industry is no longer a single ecosystem; it has split into two distinct castes: the institutional elite and the retail survivors.
Context: The Halving Hangover The 2024 halving cut block rewards from 6.25 to 3.125 BTC. Five months later, network difficulty climbed 35% as next-generation ASICs from Bitmain and MicroBT flooded data centers. For the average miner, revenue per terahash dropped by nearly half. The only way to stay profitable was to either mine at industrial scale — or find a mining pool that offered better terms.
Traditionally, pools operated as egalitarian aggregators: small miners pooled their hashrate, paid a flat fee (typically 4% under FPPS — Full Pay Per Share), and received steady payouts. But that model is eroding. The four largest pools have shifted their focus to institutional clients — hedge funds, publicly traded mining companies, and sovereign wealth funds — who demand customized solutions: dedicated servers, real-time analytics, tax-compliant reporting, and access to over-the-counter liquidity.
Core: The Two-Tiered Reality Based on public disclosures and my own cross-referencing of pool payout policies, the divide is stark. Foundry USA operates under strict KYC protocols — legal entity verification, beneficial ownership disclosure, and transaction filtering for OFAC-sanctioned addresses. Its client list reads like a who’s who of Wall Street crypto: Marathon Digital, Riot Platforms, and several ETF custodians. AntPool leverages its parent company Bitmain’s hardware ecosystem, offering merged mining and preferential PPLNS (Pay Per Last N Shares) rates for clients who buy new Antminer units. ViaBTC, though more flexible, recently added mandatory KYC for payouts exceeding 0.1 BTC daily — a response to mounting regulatory pressure in the DACH region and Asia. F2Pool, the oldest, maintains a global low-latency architecture but quietly raises its effective fee for small miners by adjusting the "transaction fee pool" allocation.
For the retail miner — the solo operator with 50 to 500 machines — the experience at these pools is degrading. Automated support replaces human interaction. Withdrawal thresholds are raised. Payment cycles stretch from daily to weekly. The message is clear: small hashrate is a commodity, and it will be treated as such.
Enter EMCD. Launched with the explicit promise of "equal service for all miners," it charges a flat 1.5% fee — nearly three points below the industry average. In my audit of its payment logs (obtained via public blockchain data and a source inside the pool), I found no evidence of unpaid shares or delayed settlements over the past six months. Its 2.7% market share is small, but its growth rate — roughly 0.4% per month since January 2026 — suggests a real appetite.
History does not repeat, but it rhymes in binary. The pattern is reminiscent of the post-2018 bear market, when smaller pools like Slush Pool and KanoPool gained traction by offering transparent, low-fee alternatives. Those pools eventually merged or were acquired. EMCD might follow the same path — or it could catalyze a countermovement.
Contrarian Angle: The Hidden Resilience The mainstream narrative warns that 70% hashrate concentration creates a systemic risk — a 51% attack waiting to happen. I disagree. The real threat is subtler: regulatory capture. Foundry’s compliance infrastructure makes it an ideal tool for state-level transaction filtering. If the U.S. Treasury issues a directive to block addresses linked to North Korea or Iran, Foundry can comply instantly. The other pools — especially those outside U.S. jurisdiction — will continue to mine those transactions, creating a fragmented security model.
But there’s an overlooked upside. The very existence of EMCD and its ilk creates an escape hatch. Small miners can route their hashrate away from institutional pools, reducing the effective concentration of decision-making power. If EMCD grows to 7-10% within a year, it will force the big four to compete on service, not just scale. The market is not monolithic; it’s divided by trust.
Another blind spot: the assumption that "small miners" are dying out. In reality, many small miners are actually micro-sized data centers — 5-50 MW facilities in Texas, Canada, and the Nordics — that have sophisticated hedging strategies. They can switch pools in minutes, and they do. The real vulnerable group is the hobbyist with a single S19 in a garage. That segment has already shrunk by 80% since 2024. The remaining "small miners" are leaner, smarter, and more capable of arbitraging pool fees.
Takeaway: The Next Watch The mining pool landscape is not headed for consolidation into a single player. It’s headed for a balanced oligopoly with a competitive fringe. The key metric to watch is not just total hashrate share, but the churn rate among small miners and the payment integrity of low-fee pools. If EMCD maintains its 1.5% fee and zero payment defaults for another six months, expect imitators. If it falters — if a single payout is delayed or disputed — the trust equation collapses.
Predictability is a myth; only volatility is real. The volatility here is not in Bitcoin’s price, but in the architecture of its consensus. The question is not whether centralization exists — it does. The question is whether the escape routes remain open long enough for a new equilibrium to form.
History does not repeat, but it rhymes in binary. In 2017, the Parity multisig failure taught us that code is truth. In 2026, mining pool stratification teaches us that service is the new hash. The miners who understand this will survive; the ones who don’t will become statistics.