Hut 8: Falling Knife or Falling Opportunity?

Hut 8: Falling Knife or Falling Opportunity?

I must admit that I've been puzzled by the market's reaction to Hut 8's second-quarter results. The company reported revenue nearly doubling year-over-year, adjusted earnings that beat consensus by 48%, and expanded its contracted AI infrastructure pipeline to nearly 950 megawatts with $26.6 billion in aggregate deal value. And yet the stock has been battered, shedding roughly 26% over the past three weeks to trade near $81, down from a 52-week high of $140.80.

What more do investors want?

The market has arguably baked in a doomsday narrative around Hut 8's transition from Bitcoin mining to a power-first, AI-focused infrastructure platform, even though the core commercial commitments that powered the stock's 76% year-to-date gain haven't cracked. That disconnect between price and the contracted revenue floor is where the contrarian setup starts.

The Q2 Miss Wasn't About the AI Thesis

Hut 8's Q2 2026 revenue came in at $74.9 million, up 81% year-over-year but $0.5 million shy of the $79.8 million consensus. The GAAP net loss was $177 million — a stark swing from $137.5 million in net income during the comparable prior-year quarter. On the surface, those numbers look ugly. But the loss was dominated by $138.6 million in mostly unrealized mark-to-market losses on digital assets, a volatility artifact that has nothing to do with the AI infrastructure business that investors are actually buying.

Stripped of crypto mark-to-market noise, adjusted EBITDA (earnings before interest, taxes, depreciation, and amortization — a rough cash-earnings proxy) climbed to $10.4 million from $4.2 million a year ago. That's not profitability in any meaningful sense for a $10 billion company. But it's the right direction for a business still in heavy buildout mode, and it beats what the Street was forecasting.

This matters because the market punished HUT as if something structural broke. It didn't. The revenue miss was a rounding error, and the headline loss was accounting noise from Bitcoin price swings — not an AI demand problem. The company's former mining operations were spun off into a separately traded subsidiary called American Bitcoin Corp. back in March 2025. Hut 8 now focuses exclusively on AI data centers.

The Contracted Pipeline Is the Real Story

Here's the data point that should anchor your thinking: Hut 8 now has 949 megawatts of contracted IT capacity across two AI data center campuses — Beacon Point in Texas and River Bend — backed by investment-grade counterparties. The expected aggregate base-term contract value is $26.6 billion, with more than $1.75 billion in expected average annual net operating income.

To put that in perspective, roughly one year ago, Hut 8 had zero contracted AI revenue. The Beacon Point campus alone has a base-term contract value of approximately $19.6 billion and expected annual NOI of $1.3 billion, with renewal options pushing potential campus-level contract value to $50.2 billion. A second 352-megawatt lease at Beacon Point, signed with the same high-investment-grade tenant as Phase 1, closed after quarter-end and brings the full 1 gigawatt of utility capacity online.

These are triple-net, take-or-pay leases. That means the tenants pay regardless of utilization. If the customer's AI workloads underperform, Hut 8 still gets paid. That's not a speculative narrative — it's a structural floor.

Beyond what's already contracted, the development pipeline totals approximately 8,660 megawatts of utility capacity. Of that, 1,330 megawatts are under construction, 1,880 megawatts are under exclusivity, and 5,400 megawatts are under diligence. The pipeline is deep and the model appears repeatable, as the Beacon Point Phase 2 lease demonstrated.

Where the Bear Case Lives

The bear case isn't that the contracts are fake — they're real, with investment-grade counterparties. The bear case is timing and leverage.

Independent analysts do not expect material AI revenue to be realized until the second quarter of 2027. That's at least 10 months away. Between now and then, Hut 8 continues to burn through capital. Total debt sits at $8.2 billion against $233.6 million in cash, giving a debt-to-equity ratio of 435.5%. Free cash flow for the trailing twelve months was negative $1.3 billion, with capital expenditures of $775.8 million.

The company did close $7.5 billion in non-recourse, investment-grade project financing in a single quarter — $4.25 billion for Beacon Point Phase 1 (rated Baa2) and $3.25 billion for River Bend. That financing doesn't require Hut 8's parent company to make good if the projects fail, which is important. And the company carries no general recourse debt at the parent level following the conversion of a $150 million convertible note.

But the balance sheet is undeniably stretched. ROIC is negative 14.4%, and ROE is negative 44.3%. This isn't a company that's profitable today. It's a company betting that contracted future cash flows from AI infrastructure will justify the buildout cost. That's a valid bet — but it's still a bet with a long runway to validation.

Price Action Suggests Exhaustion, Not Collapse

The stock has dropped from $140.80 to $80.79, a 43% decline from its 52-week high. The 20-day decline alone is 25.9%. The RSI (relative strength index, a momentum oscillator that measures whether a stock is overbought or oversold) sits at 39, approaching oversold territory. The 50-day moving average is at $104.74, well above the current price. But the 200-day moving average — the longer-term trend line — sits at $72.59, just 10% below current levels.

The technical picture is messy but not catastrophic. The stock is below its short-term moving average, confirming the selloff is real. But it's approaching the longer-term trend support at $72-$73, and the RSI is near levels where selling exhaustion typically appears. Daily volatility at 9.7% and a 20-day volatility of 10% means the stock whipsaws hard in both directions.

The selling pressure makes sense in hindsight: the stock was up 76% for the year before earnings, then reported a GAAP loss that was wider than expected, revenue was fractionally short, and the broader tech sector has been soft. But none of that changes the $26.6 billion in contracted revenue or the 949 megawatts under lease.

The Contrarian Setup

Here's how I read it: Hut 8 is a high-risk, high-conviction contrarian buy at these levels, but the entry needs patience.

The market swung from euphoria (stock up 76% for the year) to panic (down 43% from highs) in roughly four weeks. That kind of whiplash rarely has the fundamentals to back it up. The AI infrastructure thesis didn't break in Q2. The contracted pipeline grew. Adjusted EBITDA doubled. Project financing came in at investment-grade terms. What the market punished was a rounding-error revenue miss and unrealized crypto losses — neither of which affects the structural story.

But I'm not asking investors to chase this name right now. The stock is still falling, and the 50-day moving average at $104.74 looms overhead as resistance. The better risk/reward is likely near the 200-day moving average around $72-$73, where longer-term trend support sits and selling exhaustion is more likely to have played out. That's roughly 10% below current levels.

For investors who already own HUT at higher prices, the fundamental case hasn't deteriorated. Double down near the $72-$73 support zone rather than catching a falling knife at current levels.

The Risk Trigger

I'd reassess if either of two things happens. First, if a major contract with an investment-grade counterparty is renegotiated or walked away from — that would signal the triple-net structure isn't as ironclad as the narrative suggests. Second, if project financing terms deteriorate, pushing the cost of capital up or removing the investment-grade ratings — that would break the non-recourse financing model that keeps the parent company insulated from project risk.

Until those triggers fire, the market's panic looks like an overreaction. The contracted revenue floor is real. The pipeline is deep. And the selloff hasn't priced in failure — it's just priced in impatience.

Investors who can tolerate the volatility and the timeline shouldn't let this chance go to waste. But wait for the dip, not the cliff.