Over the past seven days, three mining pools — Foundry USA, Antpool, and F2Pool — have consistently controlled 67.4% of Bitcoin’s total hashrate. This is not a spike. It is a structural reality that has been tightening since the fourth halving in April 2024, when block rewards dropped to 3.125 BTC and operational margins for smaller miners evaporated. The data from mempool.space shows a steady drift: the number of solo miners has fallen by 42% year-over-year, while the top three pools have consolidated their share by 11% in the same period. If this trend continues, by mid-2026 we will see a functional oligopoly, where any two pools colluding could theoretically execute a 51% attack — or, more subtly, censor transactions. The crypto press calls this a "market maturation." I call it a slow betrayal of the immutable promise that made Bitcoin meaningful in the first place.
To understand why this matters beyond a spreadsheet row, we need to step back into the philosophical undercurrent of 2017, when I volunteered for the Ethereum Classic community in Mexico City. I spent months translating whitepapers for Spanish-speaking newcomers, arguing that "Code is Law" wasn’t just a catchy slogan — it was a moral stance against sovereign override. Back then, the enemy was explicit: central banks, corporate servers, government censorship. The solution was distributed consensus, where no single party could rewrite history. But the architects of that vision overlooked a quiet flaw: consensus algorithms are only as decentralized as the economic incentives that sustain them. Bitcoin’s SHA-256 mining was designed to be permissionless, but it assumed that capital would remain fragmented. It did not account for the industrial scale of modern ASIC farms, cheap hydroelectric contracts in Sichuan, and the network effects of pool aggregations. The halving events, by slashing revenue, accelerated the very concentration the system was meant to resist. This is not a bug in the code; it is a bug in the incentive model — a tragedy of the commons playing out in real-time on a ledger that claims to be trustless.
The core of the matter lies in the economics of mining post-halving. Every four years, Bitcoin’s inflation rate halves, reducing the block subsidy — which historically constituted 98% of miner revenue — while transaction fees remain too low to compensate (currently averaging 0.2 BTC per block, per The Block’s dashboard). For a miner operating 1,000 S19j Pro units (around 100 PH/s), the daily revenue has dropped from roughly $8,000 in early 2024 to $4,200 today, assuming $0.05/kWh electricity. That razor-thin margin leaves no room for independent miners — they either join a large pool to smooth variance or sell hardware and exit. The pools, in turn, are run by entities with their own motivations: Foundry is owned by Digital Currency Group (DCG), Antpool by Bitmain (which also manufactures miners), and F2Pool is a Chinese entity with regulatory ties. Each pool has the technical capability to filter transactions or reorg blocks without revealing their actions immediately. The argument that pools are "just service providers" is naive — they hold the ultimate power to select which blocks extend the chain. And because mining is now a business with thin margins, the pressure to optimize revenue often overrides ideological purity. I have seen this firsthand: during my 2022 audit of failing L1 protocols, I discovered that at least two mining pools had accepted "dark connections" from large traders to prioritize their transactions, effectively acting as private sequencers. The same dynamic applies to Bitcoin, only concealed by the narrative that mining is "random" — it is not. It is a highly optimized industrial process where the largest players can, and do, coordinate informally.
The contrarian angle that the industry loves to toss back is: "Hashrate concentration is fine because pools compete, and miners can switch pools instantly." That argument holds a grain of truth — in theory, a miner can point their hashrate to a different pool in seconds, preventing long-term capture. But in practice, switching requires trust in a new pool’s payout regularity, fee structure, and uptime. Smaller pools like Poolin or ViaBTC have suffered from repeated downtime and delayed payments, making them less attractive despite lower fees. Moreover, the switching argument assumes that miners are rational actors who value decentralization over profit — a false assumption in a high-capital, low-margin industry. The real test of pool power is not whether miners can leave, but whether they will leave when it matters. In a bear market, when every satoshi counts, miners stick with the pools that pay most reliably — which are precisely the largest ones. This creates a self-reinforcing loop: large pools attract more mining power, which gives them better luck, which attracts even more power. The system is not in equilibrium; it is in a spiral toward oligopoly. And the standard rebuttal — "Bitcoin is still the most decentralized cryptocurrency compared to PoS chains" — is a dodge. It’s like saying a patient with stage-3 cancer is healthier than one with stage-4. Yes, Bitcoin is less centralized than Ethereum post-Merge, but that threshold is meaningless when the trend lines are clear. The soul of the protocol — its claim to be unstoppable, uncensorable, and trust-minimized — erodes with each percentage point of hashrate concentration.
Where does that leave us? We chart the code, but the soul chooses the path. The path we are on leads to a Bitcoin that is secure by institutional convenience, not by distributed resistance. If the top three pools ever coordinate — even implicitly, through shared hardware suppliers or regulatory pressure — the chain’s immutability becomes a permissioned illusion. The solution is not to force miners to spin up more pools (which nobody can mandate), but to structurally redesign the incentive layer. One possibility is to increase the fee-to-subsidy ratio faster, perhaps through a soft fork that prioritizes fee revenue or introduces a minimal non-zero block reward that adjusts with difficulty. Another is to implement a "pool rotation" protocol similar to the concept of "crypto-note" styled proof-of-work, where each block forces a random subset of miners to validate without knowing which pool submitted the work. These changes require consensus, which in itself demands a level of social cohesion that Bitcoin’s governance has historically lacked. But the alternative — complacent acceptance of the oligopoly — is a slow death of the very principle that drew me, and millions of others, into this space. The question is not whether Bitcoin will survive; it will. The question is whether it will survive as the decentralized sovereign money it was meant to be, or as another centralized settlement layer controlled by a handful of industrial giants. The answer, as always, lies in the code we choose to write — and the soul we choose to keep.