The Power Company Everyone Suddenly Wants a Piece Of, Except the Stock Hasn’t Caught Up

There’s a specific kind of setup that gets value investors leaning forward: a company posts genuinely strong numbers, the stock drops anyway on a technicality buried in the footnotes, and then the people with the best information available (the CEO, sophisticated hedge funds) start buying with their own money. 

That’s happening right now in the power sector, and most retail investors are too busy chasing chip stocks to notice.

Vistra’s growth numbers were strong. The accounting noise stole the headline

Vistra Corp. (NYSE: VST) posted second-quarter adjusted EBITDA from its core, ongoing operations that jumped over 30% year over year, with every operating segment growing and its Texas power business more than doubling.

That’s the kind of growth number that should send a stock higher on its own.

Instead, the stock dropped because reported net income fell year over year, dragged down by a large unrealized loss tied to hedging contracts.

Here’s the distinction that matters: that loss is a paper accounting entry from marking hedges to market, not cash leaving the business.

Vistra locks in power prices well in advance to protect future revenue, and when market prices move against those locked-in contracts on paper, GAAP earnings take a hit even though the actual cash economics of the business didn’t change.

The market read the headline net income number and sold first, asked questions later.

What management is signaling next

Management didn’t blink. Full-year adjusted EBITDA guidance was reaffirmed at a range in the high single-digit billions, and free cash flow guidance held steady too.

When a company’s own leadership doesn’t touch guidance after a quarter that just got punished on a technicality, that’s usually a signal the underlying business is fine, and the market overreacted to noise.

The company is also leaning further into the story the market is supposed to be excited about right now: a data center power venture alongside major institutional partners, including a chipmaker whose name alone tends to move stocks.

Vistra has committed real capital to that venture, with more available if certain milestones are hit.

Translation: management is putting money behind the belief that data center demand for reliable power is a genuine, multi-year growth lever, not just a talking point on an earnings call.

Paper losses aren’t cash losses, and the market keeps forgetting that

This is worth sitting with, because it’s the entire reason the stock is cheaper than it should be right now.

Vistra hedges a huge share of its expected power output years in advance, locking in prices so the business isn’t at the mercy of volatile spot electricity markets. That’s a genuinely conservative, sensible way to run a power generation company.

The tradeoff is that hedge accounting creates quarterly noise.

When the market price of power moves one direction, the value of those locked-in hedge contracts moves the other way on paper, even though nothing about the actual cash the company will eventually collect has changed.

Investors who don’t dig past the headline net income number see a decline and sell. Investors who read the segment-level numbers see 31% growth in the metric that actually reflects operating performance.

Why the hedge book is a feature, not a bug

Nearly all of this year’s expected power generation is already locked in through hedges, with a large majority of next year’s output locked in too, and a meaningful chunk of the year after that.

That’s not a company exposed to swinging commodity prices; that’s a company that’s given itself years of visibility into its own cash flows.

The volatility you’re seeing in the stock price doesn’t match the volatility in the underlying business, and that gap is exactly what a value investor should want to see.

Owning generation capacity beats owning a chip design

Every AI story you’ve read this year eventually runs into the same wall: data centers need enormous, reliable amounts of electricity, and building new power plants or transmission lines takes years.

That bottleneck is why power producers who already own generating capacity, rather than companies still waiting in interconnection queues, are sitting on a genuine structural advantage.

Vistra owns one of the largest nuclear fleets in the country along with substantial natural gas generation, which means it can serve large new power customers without waiting on years of permitting and construction that newer entrants face.

That’s the difference between a company that could benefit from the AI power buildout someday and one that’s already positioned to capture it now.

The data center venture is real capital, not a press release

The joint venture Vistra committed to alongside institutional partners is specifically built to supply power to data center infrastructure, with additional funding available if the venture hits growth milestones.

This isn’t a vague partnership announcement designed to get an AI mention into the earnings call. It’s dedicated capital pointed directly at the exact bottleneck (electricity supply) that every hyperscaler in the country is currently worried about.

That’s the kind of exposure that doesn’t show up cleanly in a single quarter’s numbers yet, but it’s the kind of optionality that tends to get re-rated hard once the market actually notices it.

A buyback that’s been quietly reshaping the share count for years

Vistra has been repurchasing stock aggressively for several years now, and the cumulative effect has been a meaningful reduction in total shares outstanding.

That matters because every dollar of future earnings gets divided among fewer shares, which is a genuine, durable way to compound shareholder value that doesn’t depend on the stock price cooperating in the short term.

There’s still a meaningful chunk of buyback authorization left to deploy.

Insiders and sophisticated investors are buying the dip you’re looking at right now

Here’s the part that should actually move the needle for you. Vistra’s CEO bought shares with his own money after the stock had already fallen sharply over the prior year, at a price close to where the recent lows sat.

That’s a chief executive deciding the market got the valuation wrong.

At the same time, several large, sophisticated investors either initiated or significantly built positions in the stock during the same window the stock was under pressure.

When you see a CEO buying with personal cash and multiple large funds building positions during the exact period retail sentiment turned negative, that’s a signal worth weighing seriously, even though none of it guarantees anything.

Full-year adjusted EBITDA guidance sits in a wide range spanning billions of dollars, and it’s tempting to anchor on the high end and extrapolate a clean growth story from there.

Don’t. That guidance range is wide precisely because power markets can move, and because the growth venture in data center power is still ramping rather than fully contributing today.

The reaffirmed guidance tells you management is confident in the base business holding steady through the year. It does not tell you how fast the newer growth initiatives will actually convert into reported earnings, and conflating “guidance held steady” with “growth is accelerating on schedule” is a mistake.

Treat the wide range as evidence of a stable core business with real optionality attached, not as a precise growth forecast you can build a price target off of.

The balance sheet carries real leverage. Vistra operates with a meaningfully higher debt-to-equity ratio than some of its power sector peers, leaving less room for error if financing costs rise or a major growth initiative requires more capital than currently planned.

Data center demand could disappoint or arrive slower than expected. The entire bull case leans on hyperscalers continuing to sign long-term power agreements at the pace the market currently expects. Any broad pullback in AI infrastructure spending would remove the cleanest growth catalyst from the thesis.

Hedge accounting will keep creating quarterly noise. The same mechanism that just tanked the stock on a headline number that didn’t reflect actual cash performance will keep doing that periodically. If you can’t stomach a stock that occasionally drops on paper losses that reverse or roll off over time, this name will test your patience repeatedly.

Regulatory and nuclear operating risk sits underneath everything. A large nuclear fleet comes with strict regulatory oversight and real operational risk. Any unplanned outage or regulatory intervention at a major facility would directly hit the generation numbers this whole thesis depends on.

This is a company posting genuinely strong operating growth, sitting on a hedge book that gives it years of cash flow visibility, and holding a real structural position in the single biggest infrastructure bottleneck the AI buildout is running into.

The stock got hit on an accounting mechanic that doesn’t reflect the actual health of the business, and the people with the best information (the CEO, large institutional holders) responded by buying, not selling.

The setup here is straightforward: you’re being offered a business growing its core profitability at a healthy clip, trading meaningfully below where it sat before the hedge-driven selloff, with a buyback quietly compounding your ownership stake in the background and a data center power venture that’s still mostly optionality rather than priced-in upside.

That’s a favorable combination if you’re willing to look past a noisy headline number.

The honest risk is leverage and patience. This isn’t a debt-free balance sheet, and the data center story needs actual contracts converting into actual cash flow on a reasonable timeline, not just headlines.

If you’re comfortable owning a power producer with real leverage while that story plays out over several quarters, this is worth building a position in.

If quarterly hedge accounting noise is going to make you want to sell every time headline earnings dip, size accordingly or wait for cleaner prints.

Looking to buy? The gap between operating performance and the stock’s reaction to hedge accounting noise is the opportunity here. Scale in rather than going all at once, since this stock has shown it can swing hard on headline numbers that don’t reflect underlying cash generation.

Already own it? Hold and watch two things next: whether adjusted EBITDA guidance holds through the rest of the year, and how quickly the data center power venture starts showing up in disclosed contracts or capacity commitments. Those are the real tells on whether the growth story is converting on schedule.

Main risk to respect: Leverage is real here, and it amplifies both the upside and the downside if financing costs move or a major growth initiative needs more capital than planned. Keep an eye on the balance sheet trend, not just the earnings headline.

That's our coverage for today; thanks for reading! Reply to this email with feedback or any names you want me to check out.

Best Regards,
—Noah Zelvis
Undervalued Edge