The weekly ETH purchase by BitMine, the largest publicly traded ETH whale, just collapsed by 73%.
In a quarter where they funneled $859 million into buying their own stock, the company allocated only a fraction to its signature asset. This is not a pause. This is a structural shift.
I've tracked BitMine's wallet activity since its first 100,000 ETH accumulation in 2023. Back then, the model looked clean: raise equity, buy ETH, stake 85% of it, collect yield. The narrative was a perfect echo of MicroStrategy's BTC playbook. But the numbers now tell a different story—one of a company bleeding cash and masking dilution with repurchases.
Context: The 5% Ceiling and the Broken Lever
BitMine holds ~577,700 ETH, targeting 5% of the circulating supply (1.207 billion ETH). They have reached ~4.79%, almost there. But the road to this milestone has been anything but smooth.
- Q2 2025 net loss: $83.6 million (despite $45.7 million in staking revenue).
- Derivatives loss: $92.1 million in the same quarter—more than double staking income.
- Shares outstanding doubled year-over-year (from ~120 million to ~240 million), diluting every existing holder.
The company's response? A $4 billion stock buyback program. In the most recent week, they spent $859 million on buybacks vs. only ~$150 million on ETH purchases. That's a 5.7x ratio favoring their own equity.
Chairman Thomas 'Tom' Lee's guidance—prioritize buybacks over ETH accumulation—is the smoking gun. When the captain of the ship signals that his own stock is a better buy than ETH, the market should listen.
Core: The On-Chain Forensics of a Slow-Motion Unwind
Let me be precise: this is not a fire sale. BitMine has not sold a single ETH. But the marginal buyer has stepped away.
The data doesn't lie: - Weekly ETH purchase run-rate has fallen from ~20,000 ETH to ~5,400 ETH—the lowest since the program began. - Staking yield (2.67% APR) covers only 55% of the quarterly net loss. - The $92 million derivatives loss suggests unhedged short volatility bets or leveraged positions gone wrong. As someone who tracked the 2020 Curve treasury drain in real time, I recognize the pattern of mounting operational risk hidden inside a 'simple' staking model.
Most analysts miss the core math: BitMine's business is essentially a leveraged carry trade—borrow via equity at near-zero cost (dilution), buy ETH, stake it, earn 2.67%. But the 'cost' of equity is not zero. When shares double, the ETH-per-share ratio halves. The buyback barely offsets this dilution (only $859M vs. $12B+ market cap).
Volume spikes lie; liquidity flows tell the truth. The flow here is clear: BitMine is using stock proceeds to buy ETH, but the net ETH-per-share is declining. The 'ETH absorber' narrative is a mirage.
Contrarian: The 'Passive Whale' Trap
Everyone is fixated on the '5% target achieved' milestone. The contrarian angle is that BitMine is about to become a passive whale—a large holder that stops accumulating and becomes a potential future seller.
Why this matters: - Once the 5% target is reached, the company has no stated commitment to buy more. The narrative engine (institutional accumulation) stalls. - The stock buyback signals management's belief that BMNR is undervalued relative to ETH. This is a bearish signal for ETH marginal demand. - If BitMine ever needs to sell ETH to cover losses (derivatives margin calls, regulatory actions, or simply a run on the stock), the ~4.8% supply held by one entity could overwhelm the market. Remember the 2022 Terra collapse? Concentration kills.
"We don't buy narratives; we buy volume and liquidity flows." The flow here shows a company buying its own stock, not ETH. The 'MicroStrategy for ETH' analogy fails because MicroStrategy never diluted this aggressively and never lost money on derivatives.
Takeaway: Watch the Staking Withdrawal Queue, Not the Press Releases
BitMine has staked 85% of its ETH (4917k ETH). This locks those coins from market circulation. If the company ever needs to unstake to raise capital, the ETH withdrawal queue (currently ~20 days) will act as a warning signal.
My forward-looking judgment: The market has already priced in BitMine's continued accumulation. When that accumulation stops (likely within 2-4 weeks), the marginal buyer disappears. ETH short-term price action will reflect this absence.
For BMNR holders: sell the stock, buy ETH directly. You get cleaner exposure without the dilution and derivatives drag.
Speed is safety when the exploit is already live. The exploit here is not a smart contract bug—it's a business model that depends on infinite equity supply. When that door closes, the house of cards collapses.