NuScale Is Down 83%. It Still Isn't Cheap.

NuScale (SMR) is down about 83% from the $57.42 it hit in late 2025, sitting near $9.70 and close to the bottom of its 52-week range. The version of the story running in a few "stocks to buy with $2,000" lists this week is that a name this beaten up in the nuclear power theme must be a bargain.

The only problem is that the drop corrected the valuation. It did not change what the company actually sells.

What the 83% bought you

Here is the math the crash left standing. NuScale's market cap is about $4.17 billion. In the last twelve months the company sold roughly $10.7 million of product and engineering work. Divide one by the other and you get about 390 times trailing sales. That is not a normal "is this cheap?" multiple — it is a diagnostic. It tells you the entire $4.2 billion price is a wager on revenue that is not in the books yet.

And the revenue that is in the books is shrinking. Q2 sales were $75,000, down about 99% from $8.05 million a year earlier. Part of that is a one-time thing: the company finished a large engineering package for a Romania project and had nothing of comparable size to bill against in the quarter. But the forward story is the issue — there is no signed, priced order sitting behind the next few years of sales.

Meanwhile the balance sheet is telling the cash story, in the company's favor, in a limited way. It ended the second quarter with about $1.9 billion in cash and investments, up roughly $900 million from the prior quarter, and that jump was mostly the company selling stock — about $984.5 million of net equity proceeds in the first half. That is a real runway. It is not a business. It is what lets NuScale keep building toward a reactor it hopes to sell. The cost of that runway is dilution: Class A shares rose from about 318.5 million at the end of 2025 to 410.4 million by the end of June — roughly 29% more — and in August the company filed to sell another $750 million of shares at the market's discretion. Cash burn is heavy, with operating cash flow around $776 million negative over the trailing year, and the share count keeps growing to fund it.

A stock can fall 83% and still be priced for a future. NuScale is that stock.

The deal that anchors the bull case

The number bulls lean on is 6 gigawatts. In September 2025, the Tennessee Valley Authority and ENTRA1 Energy announced a program for "up to 6 GW" of NuScale reactors, framed as the largest small-modular-reactor deployment in U.S. history. It sounds like an order. Read the actual status and it is something else: an agreement to collaborate on planning. There is no firm megawatt count, no price, no construction start date, and the plants are to be owned and financed by ENTRA1, a startup. In its own filing language, NuScale flags the risk of being unable to "enter into binding contracts with customers."

None of that should be read as the moat being fake. The genuine asset is real: NuScale holds the only small-modular-reactor design the U.S. nuclear regulator (the NRC) has certified for commercial deployment, and in June 2025 it cleared its uprated US460 plant design — six 77-megawatt modules, about 462 MW total. In a business where the license is the product, that certification is the closest thing NuScale has to a durable competitive edge.

But a certified design and a purchase order are different animals. The market paid for the second while the company actually has the first. That gap is what an 83% fall mostly closes at the top — it strips out the part of the price that assumed the 6 GW was already banked. It does not close the gap at the bottom, because nothing on the ground has signed it yet.

Same demand, different jobs

The clearest way to see what is happening is to look at the other end of the same theme. The demand for nuclear power is real and not hypothetical — it is why Constellation Energy, which actually generates and sells the electricity, raised its full-year earnings guidance after adding more long-term power contracts, including a 15-to-20-year deal with Walmart, and why a uranium miner like Cameco can point to multi-year delivery contracts for tens of millions of pounds of fuel. Those companies sell something today; their shares got pushed back, not halved.

NuScale sits at the end of the chain that is not invoicing yet. It sells a construction timeline. That is the whole distinction. When you buy the fuel or the power, you are paying for deliveries already made or already under contract. When you buy the reactor developer, you are paying for a future that still has to clear financing, a firm order, construction, and a decade of regulatory buildout. The crash repriced the second kind of company far harder than the first — which is exactly why "it fell 83%, so it is a value stock" does not hold. A big fall in a timeline-seller makes the bet smaller. It does not turn it into a cash-flow business.

That is also why short sellers got paid to bet against this corner of the sector — roughly $2.1 billion in profit on the three biggest small-reactor names, with about 18% of NuScale's shares out on loan at one point.

What would change the math

So what does the investment actually depend on? One thing: the "up to 6 GW" becoming a firm, priced, financed order. When ENTRA1 and TVA turn the collaboration into a signed contract with real megawatts, a price, and capital in place — and ideally when the first module is actually built and turned on — that is the moment the 390-times figure stops being a diagnostic and starts being a normal growth story. Until then, NuScale is a $4 billion company betting on a future that is several years away, and its analysts' average target (around $13.50, a "Moderate Buy" consensus) is a bet on that same future, not a value floor.

The 83% fall made the over-valuation smaller. It did not price the company into the "it sells real cash flow today" column, where a discount would actually matter. If you want that column, it is at the other end of the supply chain — the ones already selling power and fuel. If you want the reactor, you can hold NuScale at these prices knowing exactly what you own: a certified design, a real runway, and a deal that is, on paper, still just a conversation.