Vistra: Hold On A Mixed Q2 Picture

Vistra (VST) remains a Hold despite strong adjusted EBITDA growth, as declining sales and rising operating costs offset the improvement. Valuation has improved materially, with the P/E falling from 76x to 23.8x on accelerated earnings, but the price-to-book remains elevated at 15x. Growth catalysts cited include the Cogentrix acquisition, nuclear power agreements with AWS and Meta, and a new AI-focused partnership with Nvidia.

Analysis

The market is already pricing VST as an AI-power scarcity asset, but the operating profile still looks closer to a leveraged merchant generator than a clean utility compounder. That matters because the equity upside is now driven less by headline partnerships and more by whether incremental contracted megawatts show up faster than maintenance, fuel, and labor inflation. At ~15x book, the stock can re-rate quickly on proof, but it can also de-rate just as fast if the growth narrative outruns realized cash flow.

For the hyperscalers, the second-order winner is not the supplier with the loudest announcement; it is the data-center operator that locks in reliable baseload at predictable pricing. AMZN and META benefit only if VST or peers can actually deliver firm capacity on time, while NVDA’s indirect benefit is mostly through continued AI capex rather than anything company-specific here. If power delivery becomes the bottleneck, the spillover is delayed GPU deployment and slower data-center commissioning, which would be a 1-3 month sentiment headwind for the whole AI infrastructure trade.

The contrarian risk is that consensus may be overestimating how much of the AI upside VST can monetize versus how much will leak to customers, regulators, and capex. Near term, the catalyst path is disclosure: contracted MW, pricing, outage assumptions, and capex intensity. Over 6-18 months, this either becomes a durable premium multiple if utilization stays tight, or a value trap if earnings quality proves cyclical and the book multiple contracts.

Key Decisions for Investors

• Keep VST on a hold/underweight until management quantifies incremental contracted MW and margin impact; if the stock rallies further without hard contract disclosures, fade strength with a 3-6 month put spread.
• Pair trade: long CEG / short VST over the next 1-3 months, expressing the view that cleaner nuclear cash flow and more transparent contracted exposure deserve a higher-quality multiple than VST's more narrative-driven rerate.
• Use AMZN and META as the cleaner beneficiaries of reliable power availability only after evidence of firm supply terms; otherwise, treat them as exposed to higher data-center power costs and possible rollout delays.
• Set an alert for any VST guidance revision on opex, plant availability, or capex; a single reliability miss would be the fastest falsifier for the AI-premium thesis and could compress the multiple sharply.
• Do not chase NVDA on the basis of this partnership alone; any benefit is second-order and likely already embedded in broader AI capex expectations.