VST: EBITDA surged, but cash growth faces a tighter bridge

The 60-second read

Five things to know

1. Generation drove the step-up. Texas and East supplied $703M of the $797M quarterly ongoing EBITDA increase.
2. Cash growth trails earnings growth. 2027 midpoint growth is 10.96% for ongoing EBITDA and 4.44% for ongoing cash before growth.
3. The acquisition changes the balance sheet. Cogentrix cash, assumed debt and preferred redemption mean September cash is not current excess cash.
4. GAAP marks and impairments need separate treatment. A $559M unrealized hedge gain coexists with a $269M Moss Landing impairment.
5. Valuation depends on what cash survives. Normalized taxes, growth capital and noncommon claims support a $130 base value versus the $133.92 close.

Three scenarios, one probability-weighted number

Analyst twelve-month scenarios; no assumed repurchases; high uncertainty.

Signposts: what would change our mind

Review dates are our monitoring schedule, not issuer-announced earnings dates. Thresholds are analyst criteria and must be graded next time.

The rally asks a cash-flow question

Vistra produced a strong operating quarter, and the share price responded. The November 9 close was $133.92. The stock rose 10.55% from its pre-release close on November 6, then reached a cumulative 18.63% gain by November 9. These returns are computed from successive closing prices. The provider's daily change field compares that day's open with its close, so it is not our earnings-reaction measure.

Our call is HOLD, with conviction 3 out of 5, High uncertainty, and a $130 base fair value. A probability-weighted $128.50 is slightly lower. These are analytical estimates, not price targets supplied by management. The business has become more valuable through stronger generation earnings and added capacity, but a larger enterprise can still leave common shareholders with a smaller cash cushion than a headline EBITDA comparison suggests. We require a 25% discount to base value, or about $97.50, before a price-only upgrade to BUY under unchanged assumptions. A stronger operating case could change that threshold.

The market-capitalization calculation uses 339,394,683 common shares outstanding on November 2, the actual cover-page count. Multiplying by the November 9 close gives $45.452 billion. The provider's $44.948 billion market capitalization uses a different denominator, so we do not mix it with the primary share count. The cover count also captures the acquisition-related share issuance more faithfully than the September quarter's weighted-average diluted EPS denominator.

The saved market history spans more than fifteen months. Our framework asks what cash reaches the common equity after those requirements. It does not assume the recent rally must reverse.

Primary source: Vistra Q3 2026 Form 10-Q. Market observations use the saved FMP snapshot dated November 9, 2026.

The print: growth survived the hedge filter

Revenue rose 27.68% to $6.347 billion from $4.971 billion. GAAP operating income increased to $1.706 billion, while net income attributable to Vistra reached $1.140 billion and diluted GAAP EPS reached $3.23. These are the three-month columns ending September 30, 2026 and 2025, not the adjacent nine-month columns. Common shareholders' income was $1.092 billion after $48 million of cumulative preferred dividends. That distinction matters when connecting net income with per-share earnings.

The stronger underlying result is visible in ongoing adjusted EBITDA, which increased 50.41% to $2.378 billion. Management attributes the improvement to higher realized energy and capacity prices, a full quarter of the Lotus plants, and Martin Lake Unit 1 returning to service. These drivers include market conditions, acquired capacity and an operating recovery. Calling the entire increase organic would overstate what existing assets achieved on a comparable basis. Cogentrix closed after quarter-end, so it is not the reason the reported September quarter grew.

The cash-flow confirmation is encouraging. Nine-month operating cash flow reached $4.330 billion, compared with $2.638 billion. All-in capital expenditures, including nuclear fuel purchases and long-term service agreement prepayments, were $2.319 billion. Their difference is $2.011 billion of cash after capital spending. That is a useful observed measure, but it precedes acquisitions, dividends and repurchases, and it is not the company's adjusted cash-before-growth measure. Keeping the labels separate prevents an attractive operating quarter from becoming an overstated distribution story.

FMP reports revenue consensus of $6.031 billion, implying a computed 5.23% revenue beat. Its EPS actual of $3.31 does not equal the filing's $3.23 GAAP diluted EPS, and the snapshot does not establish a compatible adjustment basis for its $2.98 estimate. We therefore do not advertise an adjusted-EPS beat or compare GAAP EPS with that estimate. This is a source limitation with a practical consequence: the episode's central claim rests on the primary operating and cash figures, rather than an ambiguous consensus surprise.

Primary source: Vistra Q3 2026 Form 10-Q. Market observations use the saved FMP snapshot dated November 9, 2026.

Texas and East carried the earnings gain

Texas contributed $1.227 billion of quarterly adjusted EBITDA, versus $784 million a year earlier. East contributed $979 million, versus $719 million. Together those two generation regions account for $703 million of the $797 million improvement in ongoing adjusted EBITDA, or 88.21% of the gain. Retail improved by $61 million, West by $13 million, and Corporate and Other by $20 million. This identifies the main economic engine without assigning an unsupported amount to each individual plant or commercial contract.

Retail's $98 million quarter was better than the prior $37 million, but its nine-month adjusted EBITDA declined to $939 million from $977 million. The quarter therefore does not establish a durable retail acceleration by itself. A business that sells electricity and also owns generation has offsets across weather, load and wholesale prices. Consolidation captures those offsets; a stand-alone retail margin cannot be extrapolated to the whole enterprise. Watch generation and retail together when testing resilience through a less favorable weather period.

Segment revenue requires special care. The reported segment totals include intersegment transactions and unrealized intersegment marks. Adding Retail's $4.814 billion, Texas's $2.855 billion, East's $1.830 billion and West's $133 million would overstate consolidated revenue unless Corporate and Other eliminations are retained. The external customer-revenue chart below deliberately uses revenue from contracts with customers: $4.338 billion for Retail, $253 million for Texas, $1.253 billion for East and $146 million for West. It is a different measure from total segment revenue and is labeled accordingly.

Management reported commercial availability of nearly 95% during a very hot quarter and disclosed new demand records in ERCOT and PJM. That supports the operating explanation, but availability is not a guarantee against forced outages. The completed 405 MW Pulaski solar facility and 100 MW Deer Creek solar-and-storage facility expand the operating portfolio. The 207 MW Odessa agreement runs for twenty years; its economics are not fully disclosed here. We assign no separate premium to an undisclosed contract margin, and do not count announced demand twice in the forecast.

Primary source: Vistra Q3 2026 Form 10-Q. Market observations use the saved FMP snapshot dated November 9, 2026.

Separate operating profit from marks and damage

The $559 million unrealized commodity hedge gain is material relative to $1.140 billion of net income. It represents the valuation of positions expected to settle in future years, not cash earned from delivering this quarter's electricity. Removing that gain from adjusted EBITDA is appropriate for comparing current operating performance. It would be equally wrong to remove unrealized losses in weak periods but keep unrealized gains in strong periods. The analysis applies the same rule in both directions.

The quarterly reconciliation starts with $1.140 billion of net income, adds $356 million of income tax, $254 million of interest and related charges, and $835 million of depreciation and amortization, reaching $2.585 billion of EBITDA before adjustments. The $835 million includes nuclear fuel amortization in addition to the depreciation line shown on the income statement. Using only the $711 million income-statement line would fail to reproduce the company reconciliation. The remaining adjustments lead to consolidated adjusted EBITDA of $2.360 billion; removing the negative $18 million Asset Closure result produces ongoing adjusted EBITDA of $2.378 billion.

The $269 million Moss Landing 350 impairment is noncash in the current quarter, but it is not economically meaningless. The company used a probability-weighted discounted cash-flow assessment because the timing of a possible restart remains uncertain. The battery facility's carrying value fell to about $37 million. Removing the impairment for near-term operating comparability does not restore the lost asset value, erase remediation cash costs, or establish that the facility will restart. The valuation therefore carries explicit closure spending and avoids an assumed recovery value.

The prior year also contained unusual items, including a much smaller quarterly impairment and different nuclear-decommissioning trust marks. Comparing adjusted EBITDA helps, but repeated adjustments still deserve scrutiny. Noncash compensation compensates employees with shareholder value, so our normalized cash models treat it as an economic cost even when management adds it back. Transition costs may be temporary for one acquisition, yet a serial acquisition strategy can make them recur. The appropriate question is whether cash available after all necessary spending grows, not whether each individual adjustment can be described as unusual.

Primary source: Vistra Q3 2026 release and non-GAAP reconciliations. Market observations use the saved FMP snapshot dated November 9, 2026.

The nine-check earnings-quality scorecard

The scorecard is mixed rather than uniformly clean or uniformly poor. Cash generation improved and stock compensation is modest as a percentage of revenue, but derivative marks make GAAP net income volatile. Our checks are diagnostics, not a formula that predicts fraud or a substitute for the actual disclosures. We have not computed a Beneish or Altman score from incomplete comparably defined inputs, and we do not fill missing fields with industry assumptions merely to display a number.

Nine-month cash conversion was 175.02% when operating cash flow is divided by net income attributable to Vistra. That strong ratio includes noncash depreciation, deferred taxes, derivative remeasurement and working-capital timing. It cannot be read as a recurring distribution rate. Margin deposits consumed $184 million of cash in the first nine months, less than the prior year's $361 million. An improvement in collateral timing is welcome liquidity relief, but it is not the same as a permanent improvement in unit economics.

Our receivable-days diagnostic uses average December and September net trade receivables divided by nine-month revenue, multiplied by 273 days, giving 43.38 days. Fuel-inventory days uses average fuel inventory divided by fuel, purchased power and delivery expense on the same basis, giving 15.48 days. Those definitions are deliberately disclosed: the fuel denominator contains costs that do not all pass through physical inventory, and revenue contains hedge effects. Neither ratio should be compared blindly with a retailer's turnover measure or treated as a precise collection-cycle forecast.

The GAAP quarterly tax rate was 23.80%, while guided 2027 cash taxes are just $117 million. Accounting tax expense and cash tax payments can differ substantially because of timing, credits and tax attributes. A perpetuity built from temporarily low cash taxes would overstate normalized shareholder value. We therefore show a normalized-tax valuation alongside the company's guidance. If the company can demonstrate a longer-lived cash tax advantage with primary evidence, that would support a higher cash-flow valuation without requiring an additional electricity-demand story.

Primary source: Vistra Q3 2026 Form 10-Q. Market observations use the saved FMP snapshot dated November 9, 2026.

The September balance sheet is not November cash

At September 30, Vistra reported $21.146 billion of current and long-term debt and $2.743 billion of unrestricted cash, implying $18.403 billion of reported net debt. Restricted cash, margin deposits and nuclear decommissioning investments are not ordinary cash available for distributions. In particular, the $5.266 billion investment balance largely supports obligations associated with the nuclear fleet. Subtracting that entire balance from enterprise value while ignoring the related obligations would give common shareholders an artificial windfall.

The October acquisition changes the interpretation. Cogentrix added approximately 5,500 MW and closed with about $2.2 billion of cash consideration, five million Vistra shares valued for the agreement at $185 each, and approximately $1.7 billion of assumed debt. The $185 contractual share value is not our current stock-price input. Vistra repaid the assumed debt on November 5 using cash and receivables-facility borrowings. Replacing one financing instrument with another does not make net debt disappear. We add the assumed debt once in the enterprise bridge, rather than adding both the assumption and its subsequent refinancing.

We also account for the $1 billion Series A preferred redemption. The financing-only bridge raises net debt from $18.403 billion to $23.303 billion, then adds $1.476 billion of remaining preferred claims, $622 million of forward repurchase obligations and $157 million of noncontrolling interest. Total noncommon enterprise claims are $25.558 billion. This is an illustrative bridge, not a reported November balance sheet: it excludes unreported post-quarter operating cash, repurchases, disposal proceeds and fees. A $1 billion change in those claims changes equity value by approximately $2.95 per share.

The company reports pro forma available liquidity of about $5.4 billion after the specified October-November actions, down from $8.923 billion at September. Available liquidity includes borrowing capacity, so it is not interchangeable with cash or a net-debt deduction. December still brings the Series B redemption and a $631 million final principal-and-interest payment to Nuveen. The bridge already includes the underlying preferred and forward claims; paying them swaps one claim for cash or debt rather than creating a second valuation deduction. Liquidity sufficiency and equity cheapness remain separate questions.

Primary source: Vistra Q3 2026 Form 10-Q. Market observations use the saved FMP snapshot dated November 9, 2026.

Three routes, one cautious base case

The market-implied enterprise value is $71.010 billion: $45.452 billion of computed common equity plus $25.558 billion of illustrative noncommon claims. Against the $8.025 billion consolidated 2027 adjusted EBITDA midpoint, that is 8.85 times. The consolidated midpoint is the $8.100 billion ongoing midpoint less $75 million for Asset Closure. It is not correct to use a consolidated debt bridge with an ongoing earnings number and quietly omit the closure business's cash requirements.

Route one is an unlevered cash-flow model. We start with $8.025 billion of EBITDA, deduct $1.474 billion of normalized cash tax at 25% of EBITDA less a $2.130 billion depreciation proxy, $1.824 billion of gross maintenance capital, $1 billion of assumed growth capital, $155 million of economic stock compensation and $150 million of closure cash costs above the EBITDA expense. This gives $3.422 billion of first-year FCFF. Maintenance is the guided $1.750 billion capital line plus $74 million of expected insurance proceeds, removing reliance on a recovery. These are analyst normalizations, not a company FCFF forecast.

EBITDA grows 5% annually in the model, maintenance and the depreciation proxy grow 2%, stock compensation grows 5%, and growth capital remains $1 billion. FCFF reaches $4.567 billion in 2031. Discounting at 9% with 3% terminal growth gives $66.245 billion of enterprise value and $119.88 per share after claims. Terminal value contributes 76.92% of the enterprise estimate, so discount-rate and long-run growth choices dominate. The model deliberately does not capitalize today's low cash taxes forever; it may understate the value of finite credits and tax attributes.

Route two applies 9.5 times to $8.025 billion of consolidated EBITDA, then deducts the same $25.558 billion claims, giving $149.32 per share. That multiple is our assumption, not a quoted peer average. Route three starts from $4.7 billion of guided ongoing cash before growth and subtracts closure cash, growth spending, insurance normalization, stock compensation, an assumed preferred distribution reserve and a normalized levered tax reserve. The resulting $2.108 billion of common cash capitalized at 5.5% gives $112.90 per share. Because this is an equity cash measure, debt is not subtracted again.

A 40% DCF, 40% EBITDA and 20% equity-yield blend produces $130.26, rounded to a $130 base fair value. The wide spread across methods is useful information: it exposes how much the investment case depends on taxes, growth capital and valuation multiples. Our 30% bear, 50% base and 20% bull probabilities yield $128.50. The bear assumes $7 billion of EBITDA at 7.3 times, approximately $75 per share after claims; the bull assumes $9 billion at 10.6 times, approximately $205. These are rounded twelve-month scenarios, with no assumed repurchases or extra acquisition premium.

Primary source: Vistra Q3 2026 Form 10-Q. Market observations use the saved FMP snapshot dated November 9, 2026.

Wall Street is more optimistic

The November 9 FMP snapshot aggregates twenty-two Buy ratings and three Hold ratings, with no Sell ratings, and reports a $196.76 average target, $194 median, and a $169–$255 range. These are provider-aggregated observations, not our independently collected complete set of current research reports. The target aggregate does not expose each contributor's date, so we do not describe the average as a fully refreshed post-earnings consensus. Its implied upside from $133.92 is 46.92%, while our base estimate is slightly below the price.

The dated news records are more useful for seeing revisions. A November 9 Scotiabank item records a $169 target after the release. An October 28 CIBC item records $181, down from $201, and a same-day Scotiabank item records $189, down from $207. The two October entries predate these results. We retain that distinction rather than averaging stale and fresh targets into a newly invented post-print view. One provider record names BofA in its headline and Truist in its firm field; that inconsistent row is excluded from our firm comparison.

Annual estimates supply another perspective. FMP's 2027 EBITDA average is $8.265 billion, against management's $8.100 billion ongoing midpoint and our $8.025 billion consolidated midpoint. Those definitions are not guaranteed to match; the provider does not disclose the precise adjustments used for that line. The revenue average is $23.604 billion, based on nine analysts, while its EPS average of roughly $10.03 has six contributors. Counts attach to those specific fields and should not be misrepresented as the coverage count for every estimate.

Our disagreement is about what investors retain, rather than whether electricity demand can grow. The Street's targets may recognize longer-lived tax advantages, stronger capacity pricing, more favorable contracting or lower required returns. We expose those as potential reasons to revisit our model. Conversely, a bullish label does not settle the price question when an analyst is simultaneously cutting a target. The episode uses the actual target dates and labels our own HOLD as more cautious than the provider's Buy aggregate.

Primary source: Vistra Q3 2026 Form 10-Q. Market observations use the saved FMP snapshot dated November 9, 2026.

Grade capital allocation through cash per share

Management raised and narrowed the 2026 ongoing adjusted EBITDA range to $7.1–$7.5 billion and cash-before-growth range to $4.3–$4.7 billion. With $5.639 billion of ongoing adjusted EBITDA already earned, the full-year range implies $1.461–$1.861 billion in the fourth quarter. We use that arithmetic as a checkpoint rather than treating the seasonally strong September quarter as a new quarterly run rate. Full-year guidance includes a different seasonal mix and the newly completed acquisition.

The first 2027 ranges imply midpoint growth of 10.96% in ongoing EBITDA but just 4.44% in ongoing adjusted cash before growth. Cash conversion declines from 61.64% to 58.02%. Guided net interest paid rises from $1.091 billion to $1.353 billion, and cash taxes rise from $57 million to $117 million. Other bridge items, including reclamation, remediation and investment timing, also matter. The $800 million EBITDA midpoint increase should not be described as $800 million of extra shareholder cash.

The additional $1.5 billion buyback authorization takes remaining capacity to about $2.5 billion, expected to be used by year-end 2028. An authorization is permission, not a completed reduction in shares. The company reported approximately $6.7 billion of repurchases since November 2021 and around a 30% reduction from that starting share count. Those cumulative results are substantial, yet new acquisition shares and incentive compensation can offset ongoing buybacks. Our models hold the current primary share count constant rather than granting credit for repurchases that have not happened.

We would ask management five concrete questions. How much of the 2027 cash bridge remains after the full growth-capital plan? What is the duration and realizable value of the low cash-tax position? How much liquidity must remain available for collateral under adverse commodity curves? What acquisition returns can be demonstrated after financing and incremental shares? And what evidence is needed before Moss Landing 350 can safely return? These questions connect strategic announcements to common-equity economics without assuming either flawless execution or failure.

Primary source: Vistra Q3 2026 release and guidance reconciliations. Market observations use the saved FMP snapshot dated November 9, 2026.

Ranked risks and the strongest counterarguments

The highest-ranked risk is a mismatch between the cash measure investors capitalize and the cash that can be distributed. Our normalized valuation spends money on growth and ongoing asset obligations, treats stock compensation as a cost, and avoids perpetual low cash taxes. Management's guidance remains useful because it isolates operating performance, but it is not an all-inclusive equity distribution forecast. A model that uses cash before growth as FCFF and then subtracts debt has mixed a levered cash definition with an enterprise valuation framework.

Operational reliability and commodity exposure follow closely. High demand benefits a reliable fleet, but forced outages during valuable hours can be costly. Hedges reduce exposure to adverse price changes while also limiting participation in favorable ones; they can require collateral before the ultimate contracted economics are realized. A 98% hedge percentage is neither a 98% earnings guarantee nor evidence that the remaining unhedged slice alone determines value. Counterparty, volume, basis and operational risks persist.

The strongest bull case is that scarce dispatchable capacity, new commercial agreements and the enlarged fleet produce durable cash growth beyond our 5% assumption. Tax attributes could preserve cash longer than our normalized model allows, and the capital program could earn more than its cost. In that case a $9 billion EBITDA outcome and a higher multiple are conceivable analyst scenarios. We do not require an undisclosed data-center contract to make the bull case work, and we do not assign announced megawatts a value without a margin and capital requirement.

The strongest bear case is that weather normalizes, acquired plants require more spending, financing absorbs more of the gain, and residual storage obligations worsen. The $269 million impairment is a reminder that invested capital can fail to recover its cost even when the broader industry is attractive. Helix commitments create another call on liquidity: the filing describes a $500 million initial commitment that remains callable and another $500 million contingent on commercial milestones. Our fixed growth budget is an assumption, so a larger aggregate funding requirement would reduce value. HOLD reflects a balance of credible upside and limited current margin of safety.

Primary source: Vistra Q3 2026 Form 10-Q. Market observations use the saved FMP snapshot dated November 9, 2026.

Dated signposts that can change the call

The next useful evidence is a reconciliation of the post-acquisition balance sheet with actual cash generation. The year-end report should show the consideration allocation, debt refinancing and operating contributions more clearly. That disclosure can change both enterprise claims and the earnings base. We will update the bridge from reported numbers when available; the illustrative $25.558 billion claim total is not a reason to ignore new cash information. Even a company delivering its EBITDA range can disappoint equity investors if the capital and funding demands grow faster.

December's known obligations are checkpoints rather than automatic catalysts. The Series B redemption is scheduled for December 15, and the final Nuveen payment is due December 31. Our valuation already recognizes the underlying claims. Their payment can demonstrate funding capacity, but it does not create value merely because a liability disappears from one line of the balance sheet. The exact amount of remaining unrestricted liquidity, and the conditions on borrowing capacity, matter more than the cosmetic change.

Our next earnings review is targeted for the first 2027 release covering 2026 results; the issuer has not supplied a confirmed date in the sources used here. FMP lists February 25, 2027, but that is a provider estimate and is not presented as a company-announced appointment. The analytical checks are dated to February 28 for review planning, with the actual update performed when the primary filing arrives. We do not hold the current analysis for a transcript; no call transcript was used or quoted.

An upgrade would need either a materially cheaper entry or evidence that normalized cash is sustainably above our model after growth spending and funding needs. A downgrade would follow persistent cash-conversion deterioration, an inadequately funded capital program, or a clear fall in the earnings base. These conditions are deliberately narrower than a general statement that power demand is growing. The next packet should grade these thresholds explicitly, retaining the current $130 base value so readers can see what changed in the analysis.

Primary source: Vistra Q3 2026 Form 10-Q. Market observations use the saved FMP snapshot dated November 9, 2026.

Source ledger, history and method

All financial amounts in the tables are U.S. dollars in millions unless a per-share unit, share count or percentage is explicitly shown. Values in the valuation charts are dollars per common share, not millions of dollars per share. Quarterly cash figures are computed from cumulative cash-flow statements; fourth-quarter income-statement flows are full-year less nine-month totals. We do not derive fourth-quarter EPS by subtracting rounded cumulative EPS because weighted share counts differ. Missing tagged fields remain unavailable rather than being filled from memory.

The historical packet dataset preserves all extracted inline financial facts, including dimensional segment contexts, with the source document, fact identifier, start and end dates, unit and signed value. It also preserves the exact raw annual and quarterly filings. This gives the reviewer access to the full below-the-line and segment history even where the compact web tables emphasize the decision-relevant measures. The current quarter and comparative period were reconciled to the displayed primary tables, including the distinction between consolidated and ongoing adjusted EBITDA.

The annual report identifies Deloitte & Touche LLP as auditor and reports approximately 6,390 full-time employees at December 31, 2025. Vistra is a U.S. common-stock issuer listed as VST on the NYSE; it is not an ADS structure. The analysis concerns the continuing listed parent. Completed acquisitions and the historical delisting forms examined during admission do not imply that the current common stock is delisting. The latest quarterly filing reports effective disclosure controls. Those eligibility facts do not eliminate business, valuation or operational risks.

No price, rating, analyst target, consensus estimate, quote timestamp or forecast has been silently substituted for a primary reported figure. Provider snapshots are dated and saved. Inconsistent EPS and analyst-firm records are excluded from the relevant comparisons. The price uses the November 9 close; the share count uses the November 2 filing cover; the balance sheet uses September 30 plus a clearly labeled financing bridge. Mixing those dates without a bridge would produce false precision. The most consequential author assumptions are normalized taxes, growth capital, discount rate, terminal growth and the treatment of noncommon claims.

The packet is educational research, not a personalized allocation recommendation. Its arithmetic is reproducible in the accompanying calculation file, while probabilities, multiples and the required margin of safety remain judgments. A wide scenario range is an admission of uncertainty, not a promise that every outcome falls inside it. New evidence should lead to a versioned update and a fresh review rather than an untracked change to the approved script or published packet.

Primary source: Vistra 2025 Form 10-K and saved historical filings. Market observations use the saved FMP snapshot dated November 9, 2026.