Kerrisdale Capital is short shares of Bitmine Immersion Technologies Inc (NYSEAMERICAN:BMNR), an $18 billion Ethereum-focused digit-alasset treasury (DAT) chasing a model that is on its way to extinction.
The playbook is lifted straight from Strategy (MSTR) and copied by dozens of DAT wannabes: sell stock at a premium to the value of the tokens on the balance sheet, recycle the proceeds into more crypto, and grow token-per-share in a self-reinforcing loop. That loop worked in an earlier era when scarcity and meme-like enthusiasm kept Strategy’s premium high despite constant dilution, but those conditions have vanished. Strategy’s own premium has slipped from a long-running 2–2.5x mNAV to below 1.5x after policy reversals and unpopular financings spooked investors. Into this environment BMNR has flooded the market with more than $10 billion (and counting) of new stock in just three months – roughly $170 million per day – trying to restart a flywheel that no longer spins.
The sheer velocity of BMNR’s stock issuance has turned early enthusiasm into fatigue, with investors conditioned to believe every rally will be met by more supply. While the company still reports headline growth in total ETH tokens, the pace of ETH-per-share accretion has slowed as the NAV premium narrows and the share count balloons – a trend made harder to follow after BMNR quietly dropped up-to-date share counts from its weekly press releases. The recent $365 million direct offering, trumpeted as “materially accretive” at a headline $70 price, was in reality a discounted giveaway once the value of the attached warrants is accounted for. Rather than a display of strength, the transaction was a cleverly packaged dilutive raise that sacrificed longterm credibility for short-term cash, undermining the very premium narrative BMNR is trying to sell.
The competitive landscape for DATs is undergoing radical change. The Block recently tallied more than $100 billion in planned capital raises by US-listed firms chasing similar cryptotreasury strategies this year alone. Investors have gone from a handful of choices to being overwhelmed by crypto vehicles, offered through a thicket of near-identical DATs and soon to be amplified by a tidal wave of ETFs that will make access even cheaper and more liquid. With scarcity gone, NAV premiums across the sector have collapsed; many DATs now trade at or below parity as their reflexive loops stall and their models crumble.
Our thesis is not a bet against Ethereum itself, but rather one against the notion that investors should still pay a market premium for it. If you want ETH, just buy it directly, stake it with minimal friction, or hold it in one of the rapidly proliferating ETFs. BMNR’s pitch rests on the idea that it can deliver more than the token alone, yet the strategy is generic, the competition is mushrooming, disclosures have grown opaque, ETH-per-share has slowed, and capital raises promoted as “premium” are in reality dilutive. Against this backdrop, BMNR’s premium is destined to sink further to those of its already rekt peers.
Investment Overview
BMNR is a Las Vegas-based digital-asset company that has rapidly reinvented itself from a thinly capitalized immersion-cooled Bitcoin miner into an institutionally backed, Ethereumfocused corporate treasury (DAT). In mid-2025 the company executed a reverse stock split, uplisted to the NYSE and closed a transformative $250 million private placement led by prominent crypto investors and anchored by Tom Lee of Fundstrat, who became executive chairman and the public face of the pivot. Legacy Bitcoin-mining operations generated just $3.3 million in revenue and negative EBITDA in 2024 and are no longer material to the company’s financial profile.
The new strategy channels nearly all fresh capital into acquiring Ethereum and ETH-equivalent assets for its balance sheet. BMNR presents this as a strategic “pick-and-shovel” bet on the future of finance: as the so-called stable-coin wars unfold, with banks, tech firms and consumer brands issuing competing tokens, management argues that enduring value will accrue to the Ethereum settlement layer they all rely on. This thesis has been buoyed by the recently passed “Genius Act” and an expected executive order opening the $9 trillion U.S. retirement market to crypto allocations, developments that have already helped push total crypto-asset market value above $4 trillion.
The pivot has unfolded at remarkable speed but at considerable cost to shareholders. Starting from essentially zero ETH in July 2025, BMNR now reports holdings above 2.8 million ETH tokens. To finance these purchases, we estimate the company has issued over 240 million shares through its at-the-market (ATM) program, raising over $10 billion in gross proceeds and retaining capacity to raise another $14 billion. Continued ATM issuance, future direct equity offerings, and potential warrant exercises will further expand the share count and, judging by recent investor sentiment, will likely pressure the stock.
A recent $365 million registered-direct sale at $70 per share was marketed as a “premium” equity offering despite the heavy inclusion of warrants that offset all of the claimed accretion (more on this later in the report). We think this speaks to a management team increasingly concerned with a flagging NAV premium and stalled ETH-per-share growth.
We are constructive on ETH’s long-term role in powering DeFi infrastructure but don’t think BMNR will be able to sustain a premium over the value of its token holdings. The company is trying to copy Strategy’s once-successful playbook at a time when that very model is faltering, weighed down by questionable financings, eroded trust in management, intensifying competition, and the weight of compressing premiums.
Copying MicroStrategy as the Original DAT Falters
BMNR highlights five levers for compounding ETH-per-share: issuing stock at a premium to the dollar value of ETH held per share, using equity-linked structures to monetize volatility, reinvesting modest operating cash flow into more ETH, earn staking rewards, and opportunistically acquiring subscale or failed DATs at or below NAV parity. The core of the strategy revolves around the same flywheel that powers all DATs – a rich multiple to net asset value (mNAV) lets a company issue equity well above the value of its underlying tokens, quickly recycle the proceeds into more crypto, and grow tokens-per-share.
The problem is that this model, originally run by Strategy, has lately been showing troubling cracks. After trading comfortably above 2.0x–2.5x mNAV for most of the year, Strategy’s premium has slumped in recent months to 1.4x. The slide followed a series of criticized preferred financings, the proliferation of inexpensive and simple BTC vehicles, and an abrupt August 2025 policy shift – just two weeks after reporting 2Q earnings – when the company scrapped its pledge to issue equity only above 2.5x mNAV in favor of a looser “opportunistic” approach.
The reversal rattled investors and the ensuing decline in mNAV has spurred talk that the DAT model itself may be stuck in a “death spiral.” The shift in sentiment has reverberated across the DAT space. Despite a powerful rally in Bitcoin, Ethereum, and Solana since mid-year, DATs have seen their mNAVs collapse.
Investors have become so skeptical of the model that even routine PIPE registration statements have recently caused DAT shares to plummet, reflecting dilution and impending share sale concern. Newly launched ETH treasury peers like EthZilla (ETHZ) and SharpLink Gaming (SBET) are already trading at or below 1.0x, with ETHZ recently announcing the abrupt resignation of its CEO. In desperation, SharpLink and EthZilla have resorted to buyback programs to try to salvage their premiums, while Blockchain Technology Consensus Solutions (BTCS) has turned to gimmicks like a “Bividend” and “loyalty payments” to reignite interest. The lesson should be clear: NAV premiums are fragile, built on trust, disciplined capital allocation, and scarcity. Once they crumble, they can be difficult to rebuild.
Read the full report here by Kerrisdale Capital
