Vistra
This is the archived full research. Read the story
An American power producer: it runs gas, coal and nuclear plants, sells the electricity to 5 million homes and businesses, and uses each side to hedge the other.
The scores
Growth3/4 ↑
Ongoing Operations Adjusted EBITDA is up over 30% YoY, but a material part is Energy Harbor/Lotus/Cogentrix M&A and hedge/price timing, not pure organic volume growth.
Business quality3/4 →
Irreplaceable licensed nuclear/coal plus an integrated retail hedge command real premiums, but untested by the emerging behind-the-meter substitution wave.
Financial position3/4 →
Investment-grade at two of three agencies with $6.3B of liquidity, but net debt rose about $3B in six months as acquisitions and buybacks ran together.
Execution record3/4 →
Guidance held through an ERCOT pullback and the buyback target was hit early, but the post-2021 restructured entity's clean record is still short.
Per share capture4/4 →
171 million shares retired against one small Cogentrix-related issuance - buybacks dominate the share-count trend by a wide margin.
Potential3/4 ↑
A large upside on observable steps - Batch Zero, Cogentrix, uprates - with one piece, the Meta PPA, already committed and in progress.
Risk profile3/4 →
A real regulatory dependency (Batch Zero, PJM tariff filings) and hedge-driven earnings volatility, but diversified across ERCOT/PJM/other ISOs and survivable if either resolves unfavorably.
Valuation4/4 →
About 30% under a base-case fair value of about $173 at an 8% required return: cheap, past the 20% margin of safety.
Evidence quality3/4 →
Central claims are mostly primary filings and transcript, cross-checked externally, but the cohort is thin and the TTM EBITDA figure is hand-computed, not filed.
The Verdict
We rate Vistra Attractive. It runs gas, coal and nuclear plants that nobody can build again soon and sells their power wholesale and to its own retail customers, a solid business. Hedges swing reported profit, but the cash business grew more than 50% in the September quarter. Valued on guided profit, with debt paid first, the shares are worth about $173 against $135. The cushion is thin: part of the growth came from purchases, Texas power prices are softening and new demand may route around the grid. The 2027 range holding is the test.
The long caseAI data centres, factories and a growing population are raising power demand in Texas and the eastern grid faster than in decades, and Vistra's existing plants sell into it.
How we weighed it
The price against our value
In its favour
Price at 78% of our value
The price is still under four-fifths of our value on the guided profit after a quiet rise, with growth slowing and debt paid first, though the cushion has thinned.
The business
In its favour
Solid
Plants nobody can build again soon, supplying power both wholesale and to its own retail customers.
Growth against what the price assumes
Cuts both ways
The guide rose, but partly from plants bought and contracted; new signed data-centre demand is still small.
What could go wrong
Against it
Data centres may make their own power, Texas prices are softening, borrowing rose with buybacks and the grid queue is paused.
Against competitors
Cuts both ways
Rivals' reported profit is distorted by hedges, so the comparison proves little either way.
How its group is behaving
Cuts both ways
Power producers move together on AI demand news; its own fact is its plant base and retail arm.
The latest news and report
In its favour
The latest report passed our test with a higher range; the new signed demand behind it is small.
The evidence under this - 9 claims
The call confirmed the higher range includes the Cogentrix plants closed on 30 October and the Meta PPA, which the old aim excluded, so some of the rise is acquired or contracted rather than a higher view of the old fleet; the 2027 interest expense in the guidance reconciliation is $1,430M, against $829M in the first nine months of 2026 (S24).
- 2026-11-06editedthe 6 November release initiated 2027 guidance above the $7.4-7.8B aim
- 2026-11-09editedThe Q3 call answered the open question: the 2027 range includes Cogentrix and the Meta Perry PPA, with no split given
The comparison leans on a vendor-reported five-year prior-period median (11.51x) that the pack itself does not fully audit the way it audits current-period figures, and on a hand-computed TTM Adjusted EBITDA (C09) rather than a single disclosed number.
- 2026-11-06editedre-struck on the Q3 trailing figure and the 5 November enterprise value: 10.0 became 8.6
- 2026-11-09editedRe-struck at the 6 November close when the fair value was added.
The vendor's separate top-level market-tier EV/EBITDA of 10.07x (S16) is close to the hand-computed Adjusted figure (C11), so not every vendor EBITDA field is equally distorted - the cohort-specific 21.1x figure appears to draw on a different, more GAAP-exposed quarter mix than the smoothed top-level ratio.
Trailing GAAP operating income carries hedge marks (C06), so the 20.6% starting margin is itself noisy, and the debt ahead of the shares makes the value sensitive: with no growth after 2026 and an 18% margin it is about $50. Low cash taxes from nuclear credits would lift it above the base case.
- 2026-11-09editedRe-struck on the upgraded method: every view read at 9%, the return on new capital set by the moat, and capital per dollar of sales read from the filings.
- 2026-11-09editedRe-struck at the 8% required return the owner chose after the roll-out screen: the market priced the median name on the roster at about 7.7%, so 9% on top of a 20% margin read most of it Unattractive.
- 2026-11-09editedRe-struck on the decaying fade the owner chose on 9 November: after the stated path each year closes 30% of the gap to 4%, and an analysts' margin is cut by their gap to the filed operating profit.
- 2026-11-10editedre-valued on the owner's rules of 10 November: the return on new capital glides to its terminal level over years six to ten
- 2026-11-13editedPrice re-read at the 12 November close of $134.51, about 78% of the unchanged base case: still inside the 20% margin of safety.
The company
Vistra's fleet is 62% gas, 20% coal, 15% nuclear and 3% solar and batteries, selling into six American wholesale power markets, mainly Texas's grid, ERCOT, and PJM in the mid-Atlantic and Midwest.
It also sells electricity and gas directly to about 5 million homes and businesses in 18 states and Washington, DC. A buyer who does not use Vistra buys from a rival generator on the same grid, or builds its own plant.
The evidence under this - 2 claims
How it makes money
It earns the gap between what it costs to run its plants and what power sells for. Owning both the plants and the customers lets it lock in prices years ahead: when wholesale prices fall, retail margins cushion the blow. The contracts that do that are revalued every quarter, which is why reported profit jumps around while cash does not.
The evidence under this - 3 claims
Where the money comes from, stream by stream
| Stream | Share of sales (FY2025) | How it earns it |
|---|---|---|
| Retail electricity in Texas (ERCOT) | 50.5% | Energy charges billed to homes and businesses through TXU Energy and other brands, per kilowatt-hour delivered. |
| Retail electricity in the Northeast and Midwest | 22.9% | The same retail service sold under several brands in the deregulated states outside Texas. |
| Wholesale generation sold into grid markets | 18.0% | Power from its gas, coal, nuclear and solar plants sold to grid operators such as ERCOT and PJM at market prices. |
| Other wholesale contracts | 6.5% | Ancillary services, auction revenue and sales to other retailers, municipalities and wholesale buyers. |
| Capacity payments | 1.3% | Net payments from grid operators, mostly PJM, for keeping plants available, after capacity Vistra buys. |
| Hedging, tax credits and other | 0.9% | Realized hedge gains and transferable nuclear tax credits, offset in 2025 by unrealized hedge losses. |
Who runs it
Jim Burke has run Vistra since 2022, after years as its finance and operations chief; Kris Moldovan, a company veteran, has run the money since the same year. On their watch Vistra bought more power plants, signed long nuclear contracts with big tech buyers and beat its own profit forecast. Insiders own under 1% of the company. Bonuses ride on adjusted profit and cash flow; long-term shares on cash flow per share and the stock against the market.
The evidence under this - 3 claims
Both bonus profit metrics exclude unrealized hedge losses and were targeted at the guidance midpoint, which Vistra beat; nothing pays on the return earned on acquisitions, the main use of growth capital.
Why it wins, and who could beat it
Why customers come back
Its six nuclear reactors, 6,448 MW licensed to run into the 2030s and beyond, cannot be copied: no large reactor is under construction anywhere in the United States, and only two were finished in the past decade.
That scarcity is why a technology company wanting steady, carbon-free power for decades came to Vistra's existing plants, at a price the finance chief called above market, instead of waiting for someone to build new ones.
The evidence under this - 3 claims
This is the only publicly quantified large signed data-center commitment across Vistra's entire ERCOT/PJM footprint; there is no comparably sized signed PPA disclosed in ERCOT specifically.
The threat
Data-centre owners are starting to build their own power. About 2 GW of off-grid generation was running by mid-2026, most of it gas turbines at xAI's sites, and Microsoft and Oracle are pursuing on-site power too. Every data centre powered that way is demand that never reaches Vistra's grid.
The evidence under this - 1 claim
Behind-the-meter capacity (about 3 GW by year-end 2026) remains a small fraction of the 12-15 GW of realistic new ERCOT load alone (C29), and cannot serve Vistra's 5-million-customer retail book or offer investment-grade counterparty risk transfer the way a PPA does (machine.threat.not_substituted).
The rivals: SPCX, BE, MSFT, NRG, CEG, TLN.
Who else does this, rival by rival
| Ticker | Rival | How far along | Priced against it |
|---|---|---|---|
| SPCX | xAI (Colossus 1 and 2 gas-turbine sites near Memphis)Priced against it: Different product | On sale | Different product |
| BE | Bloom Energy (fuel-cell systems; 2.8 GW master agreement with Oracle within $7.65B of 2026 data-center contracts)Priced against it: Different product | On sale | Different product |
| MSFT | Microsoft (announced gas-turbine-powered data-center sites)Priced against it: Different product | Announced | Different product |
| NRG | NRG EnergyPriced against it: Yes | On sale | Yes |
| CEG | Constellation EnergyPriced against it: Yes | On sale | Yes |
| TLN | Talen EnergyPriced against it: No usable figure | On sale | No usable figure |
How big it could get
How big this could get
Demand is rising after decades of flat growth. Texas's grid peaked at 83.7 GW in 2025 and above 91 GW in July 2026, and its planners point to about 152 GW by 2030.
The queue of requests to connect large new users there is reported near 410 GW, mostly data centres. Vistra's own chief executive thinks only 12 GW to 15 GW of that is realistic by 2030.
The evidence under this - 4 claims
This 12-15 GW figure is Vistra's own filtering of the queue and is not independently verifiable; ERCOT and the Public Utility Commission of Texas's own Batch Zero audit (C32) exists precisely because nobody outside the generators has published a verified conversion rate from queue to real load (see machine.prize.composed_route).
Who has signed
One large buyer. Meta signed 20-year agreements for 2,609 MW from Vistra's nuclear plants in the PJM region, starting late 2026. Everything else in the data-centre story, including all of Texas, is still in negotiation.
The evidence under this - 1 claim
This is the only publicly quantified large signed data-center commitment across Vistra's entire ERCOT/PJM footprint; there is no comparably sized signed PPA disclosed in ERCOT specifically.
The results
The numbers
Sales grew 20% over the twelve months to Sep 2026, faster than the 3.0% in the year to Dec 2025.
Operating margin is 21%, up from 18% in the year to Dec 2023.
Yearly sales growth has ranged from +3.0% to +17% over the last 5 years.
Each dollar put into the business earns 14% a year after tax, about where it was (11%) in the year to Dec 2023.
It turned 13% of sales into free cash, $2.61B over the last twelve months, after spending 15% of sales on buildings and equipment.
It owes $18.4B more than the cash it holds, about 3.2 years of the cash its operations bring in.
The share count fell 30% in 4.8 years, and pay in shares costs 0.6% of sales.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-11-10. Fiscal years end in December.
| Dec 2021 | Dec 2022 | Dec 2023 | Dec 2024 | Dec 2025 | 12 mo to Sep 2026 | |
|---|---|---|---|---|---|---|
| Size and growth | ||||||
| Sales | $12.1B | $13.7B | $14.8B | $17.2B | $17.7B | $20.6B |
| Sales growth | +5.5% | +14% | +7.7% | +17% | +3.0% | +20% |
| Profit | ||||||
| Operating margin | −13% | −8.6% | 18% | 24% | 11% | 21% |
| Net margin | −11% | −8.9% | 10% | 15% | 5.3% | 13% |
| What growth costs, per sales dollar | ||||||
| Selling, general and admin | 8.6% | 8.7% | 8.9% | 9.3% | 9.7% | 8.4% |
| Pay in shares | 0.4% | 0.5% | 0.5% | 0.6% | 0.6% | 0.6% |
| Capital spending | 8.6% | 9.5% | 11% | 12% | 16% | 15% |
| Cash | ||||||
| Cash from operations | −$206M | $485M | $5.45B | $4.56B | $4.07B | $5.76B |
| Free cash flow | −$1.24B | −$816M | $3.78B | $2.48B | $1.32B | $2.61B |
| Balance sheet | ||||||
| Cash and short-term investments | $1.32B | $455M | $3.48B | $1.19B | $785M | $2.74B |
| Debt | $10.7B | $12.6B | $14.4B | $16.3B | $18.8B | $21.1B |
| Net cash | −$9.41B | −$12.2B | −$10.9B | −$15.1B | −$18.1B | −$18.4B |
| Returns and shares | ||||||
| Return on capital | - | - | 11% | 16% | 7.0% | 14% |
| Shares (diluted) | 482M | 422M | 375M | 353M | 346M | 338M |
What growth costs
Running Vistra gets no cheaper as it grows: office and selling costs rise in step with sales. Growth costs capital instead, buying plants and spending more each year upgrading them, so whether growing makes it more profitable depends on what those plants earn, which the reported figures, swung by hedging gains and losses, do not yet show.
The evidence under this - 2 claims
- Revenue carries unrealized hedge marks, a $766M loss in 2025 against a $1,013M gain in 2024 (C07, C42), so every per-sales-dollar ratio moves with commodity prices as well as with costs.
- The 2025 pay plan's total-cost metric missed its threshold, on growth projects, the Martin Lake 1 repair and integration costs (S20), so costs outside SG&A did rise faster than planned.
The last report (6 Nov)
Quarter ended September 2026, filed figures.
The September 2026 quarter beat on every measure we watch. Underlying profit, the company's own measure without hedge swings, rose to $2,378M from $1,581M, and reported profit was $1,140M, helped by a $559M paper gain on hedges.
Vistra raised the middle of its 2026 forecast to $7.1B to $7.5B and gave its first 2027 forecast, $7.7B to $8.5B, above the $7.4B to $7.8B aim it had held. It added $1.5B to its buyback and signed a 20-year, 207 MW power deal in Texas. On the call it said the 2027 range includes the Cogentrix plants and the Meta deal, without saying how much each adds.
What it did to the thesisIt passed the test we set, with a caveat: the higher 2027 range includes bought and newly contracted plants the old aim left out. New signed demand is small, and debt stays moderate, so the view and conviction hold.
The quarter in detail
Sales growth from the previous quarter, filed figures.
Earnings per share $3.23 against $2.98 expected.
The evidence under this - 5 claims
Nine-month Ongoing Operations Adjusted EBITDA was $5,639M, so the range implies $1,461M to $1,861M for the fourth quarter against $2,378M in the third, and the fourth quarter carries Cogentrix for two months (S24).
- 2026-11-06editedthe 6 November release narrowed 2026 guidance and raised its midpoint
The call confirmed the higher range includes the Cogentrix plants closed on 30 October and the Meta PPA, which the old aim excluded, so some of the rise is acquired or contracted rather than a higher view of the old fleet; the 2027 interest expense in the guidance reconciliation is $1,430M, against $829M in the first nine months of 2026 (S24).
- 2026-11-06editedthe 6 November release initiated 2027 guidance above the $7.4-7.8B aim
- 2026-11-09editedThe Q3 call answered the open question: the 2027 range includes Cogentrix and the Meta Perry PPA, with no split given
Management gave no quantified Cogentrix contribution to 2027 and no new signed load beyond the 207 MW deal, and said ERCOT forwards are depressed, so the per-share path leans on buybacks and assets not yet contracted.
What management did with the money
It has bought back about 30% of its shares since November 2021, at an average of about $38, and has $1.2B of authority left. It has also bought Cogentrix's 10 gas plants, closed on 30 October, for about $2.3B in cash plus shares. Debt rose to $19.6B from $17.0B in six months.
The evidence under this - 4 claims
Management's own language - 'we expect to exhaust that no later than end of 2027' - is itself a target, not a commitment, and the same capital could instead go to additional debt paydown if the investment-grade ratings push takes priority, which management named as a live alternative use of cash.
- 2026-11-06editedthe 6 November release added $1.5B to the buyback authorization
- 2026-10-30editedThe acquisition closed on 30 October 2026, inside the mid-to-late 2026 window; the claim moves from forward to historical.
The price, and where we disagree
In short. Cheap. Below its own history, and about 78% of what we think the business is worth: inside our cushion, if only just.
Most coverage calls Vistra an AI-power play, and the standard price-to-earnings figure makes it look expensive. That figure is struck on the distorted reported profit. On underlying profit the whole company cost about 9.1 times the last year's at the 6 November close, after the shares fell about 30% from 9 October while the forecast went up; they closed at $134.51 on 12 November.
The next test is the year-end results in February 2027: whether the 2027 range holds and what signed demand has been added.
The evidence under this - 3 claims
The vendor's separate top-level market-tier EV/EBITDA of 10.07x (S16) is close to the hand-computed Adjusted figure (C11), so not every vendor EBITDA field is equally distorted - the cohort-specific 21.1x figure appears to draw on a different, more GAAP-exposed quarter mix than the smoothed top-level ratio.
The comparison leans on a vendor-reported five-year prior-period median (11.51x) that the pack itself does not fully audit the way it audits current-period figures, and on a hand-computed TTM Adjusted EBITDA (C09) rather than a single disclosed number.
- 2026-11-06editedre-struck on the Q3 trailing figure and the 5 November enterprise value: 10.0 became 8.6
- 2026-11-09editedRe-struck at the 6 November close when the fair value was added.
The call confirmed the higher range includes the Cogentrix plants closed on 30 October and the Meta PPA, which the old aim excluded, so some of the rise is acquired or contracted rather than a higher view of the old fleet; the 2027 interest expense in the guidance reconciliation is $1,430M, against $829M in the first nine months of 2026 (S24).
- 2026-11-06editedthe 6 November release initiated 2027 guidance above the $7.4-7.8B aim
- 2026-11-09editedThe Q3 call answered the open question: the 2027 range includes Cogentrix and the Meta Perry PPA, with no split given
How we value it
- Method
- Cash the business will produce, in three stages
- Required return
- 8%
- Long-run growth
- 4%
- Moat
- An average moat
| Bear | Base | Bull | |
|---|---|---|---|
| Sales growth, years 1 to 3 | 23%, 0%, 0% | 23%, 8%, 5% | 23%, 12%, 8% |
| Normal operating margin | 18% | 27% | 31% |
| Return on new capital after year ten | 8% | 12% | 15% |
| Value a share | $48.41 | $173 | $259 |
- Price
- $135
- Priced on
- 12 Nov
- Price as a share of the base value
- 78%
- Starting growth the price pays for
- 3.7%
- Yearly return the price offers on the base case
- 8.7%
- Risk check: valued at 10%, the rate for a cyclical or indebted business
- $91.08
Where each figure comes from
| Figure | Source | Where exactly |
|---|---|---|
| Sales, the last full year | Filed report | Annual report (10-K) for the period to 31 Dec 2025, filed 27 Feb 2026; the full year. |
| Operating margin today | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 9 Nov 2026; the four quarters to that date. |
| Tax rate | Our judgement | Our reading of the evidence. |
| Sales per dollar of new capital | Filed report | Annual report (10-K) for the period to 31 Dec 2025, filed 27 Feb 2026: the year's sales over its operating capital. |
| Shares | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 9 Nov 2026; the diluted share count. |
| Cash and investments | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 9 Nov 2026; the balance sheet on that date. |
| Debt | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 9 Nov 2026; the balance sheet on that date. |
| Sales growth, years 1 to 3 | Our judgement | Our reading of the evidence, set against the analysts' forecasts and the company's guidance; the base case below says where it follows them and where it departs. FMP's consensus of 5 to 10 analysts read 12.3%, 5.5%, 2.7%. |
| Normal operating margin | Our judgement | Our reading of the evidence; each case below names it. FMP's consensus, less stock pay, read 13.9%. |
| Return on new capital after year ten | Our method's rule | Set by our reading of the moat: the 8% rate with none, 12% with an average one, 15% with a strong one; the bear case earns the rate. |
| The bear and bull cases | Our judgement | Our reading of the evidence; each case below names it. |
| Price | Market price | FMP's closing price on 6 Nov 2026. |
| Preferred stock | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 9 Nov 2026; the balance sheet on that date. |
| Minority interest | Filed report | Read by hand from the filed statements; the period it was read from was not recorded. |
What each case rests on
Bear caseTexas power prices keep softening on batteries (C34) and only about 12-15 GW of the load queue is real (C29): sales stop growing after 2026, the margin falls to 18%, and new capital earns only 9%.
Base case2026 grows 23% to about $21.9B, the analysts' figure behind the raised $7.1-7.5B guide (C04); 2027 8%, matching the $7.7-8.5B range (C05); then 5%; the operating margin rises to 27% from 20.6% as the 20-year Meta nuclear contract (C20) and data-centre deals reprice the fleet.
Bull caseMore signed data-centre load beyond Meta lifts 2027 to 12% and 2028 to 8%, and the margin reaches 31%, about the analysts' 2030 figure of $7.9B of operating profit on $26.1B of sales.
What moves the base value
| Lower | Higher | |
|---|---|---|
| Growth in years 1 to 3, a quarter lower or higher | $156 | $191 |
| The normal margin, 15% lower or higher | $136 | $210 |
| Capital per dollar of growth, a quarter more or less | $163 | $179 |
| Return on new capital after year ten, at the rate or one step up | $126 | $192 |
| The rate, a point higher or lower | $124 | $254 |
| Long-run growth a point lower | $150 | - |
Value a share as growth and margin change
| Normal margin ↓ · growth → | 50% | 75% | 100% | 125% | 150% |
|---|---|---|---|---|---|
| 21.6% | $100 | $111 | $124 | $137 | $151 |
| 24.3% | $120 | $134 | $148 | $164 | $180 |
| 27% | $140 | $156 | $173 | $191 | $210 |
| 29.7% | $160 | $178 | $197 | $218 | $240 |
| 32.4% | $181 | $200 | $222 | $245 | $269 |
Growth is shown as a share of the base case's. Green: the price is at or under 80% of that value. Red: at or over 120%.
How we read the multiple
At the 6 November close, counting its debt, the whole company cost about $67B, roughly 9.1 times the last twelve months' underlying profit before interest, tax and depreciation of about $7.4B.
Valued on that profit growing as guided and then slowing to 4% over ten years, with $24.9B of debt paid first, the shares are worth about $173. At $134.51 on 12 November the price is about 78% of that: just inside our 20% margin of safety.
Against its own history. Well below its own normal on underlying profit. The reported figure makes it look dear, but that comes from paper losses on hedges, not the business.
The evidence under this - 6 claims
Trailing GAAP operating income carries hedge marks (C06), so the 20.6% starting margin is itself noisy, and the debt ahead of the shares makes the value sensitive: with no growth after 2026 and an 18% margin it is about $50. Low cash taxes from nuclear credits would lift it above the base case.
- 2026-11-09editedRe-struck on the upgraded method: every view read at 9%, the return on new capital set by the moat, and capital per dollar of sales read from the filings.
- 2026-11-09editedRe-struck at the 8% required return the owner chose after the roll-out screen: the market priced the median name on the roster at about 7.7%, so 9% on top of a 20% margin read most of it Unattractive.
- 2026-11-09editedRe-struck on the decaying fade the owner chose on 9 November: after the stated path each year closes 30% of the gap to 4%, and an analysts' margin is cut by their gap to the filed operating profit.
- 2026-11-10editedre-valued on the owner's rules of 10 November: the return on new capital glides to its terminal level over years six to ten
- 2026-11-13editedPrice re-read at the 12 November close of $134.51, about 78% of the unchanged base case: still inside the 20% margin of safety.
Net debt is the 30 June figure moved only by the events the release names, so it takes no credit for Q3 cash flow and is likely a little high; the 10-Q will replace it.
- 2026-11-06editedre-struck at the 5 November close, pro forma for Cogentrix and the preferred redemption, because the Q3 balance sheet is not yet filed
- 2026-11-09editedRe-struck at the 6 November close when the fair value was added.
This is a hand-computed TTM figure, not a single filed number - it depends on H1 2025/H1 2026 Ongoing Operations figures being genuinely comparable across two different guidance vintages and any quarterly restatement between them would move it without a corresponding filing to flag it.
- 2026-11-06editedthe Q3 release lets the trailing figure roll to September 2026
The comparison leans on a vendor-reported five-year prior-period median (11.51x) that the pack itself does not fully audit the way it audits current-period figures, and on a hand-computed TTM Adjusted EBITDA (C09) rather than a single disclosed number.
- 2026-11-06editedre-struck on the Q3 trailing figure and the 5 November enterprise value: 10.0 became 8.6
- 2026-11-09editedRe-struck at the 6 November close when the fair value was added.
The vendor's separate top-level market-tier EV/EBITDA of 10.07x (S16) is close to the hand-computed Adjusted figure (C11), so not every vendor EBITDA field is equally distorted - the cohort-specific 21.1x figure appears to draw on a different, more GAAP-exposed quarter mix than the smoothed top-level ratio.
Priced against its rivals
The competitors its own annual report names, each confirmed against the rival's filing. Measured on company value for each $1 of cash profit, cheapest first:
| Ticker | Company | company value for each $1 of cash profit |
|---|---|---|
| NRG | NRG Energy | 11.3× |
| CEG | Constellation Energy | 16.8× |
| VST | Vistra | 21.1× |
| TLN | Talen Energy held out | not priced |
The middle of the other 2 is 14.1. Vistra Corp. is dearest of 3.
Rivals' figures come from a data provider, so read the spread of the group rather than any single pair.
The rivals' figures are struck on reported profit, which the hedges distort, so the comparison proves little either way and cannot move our view.
Where we disagree with the market
An AI-power story, with a high price-to-earnings figure showing the market has already paid up for it.
The reported swings are mostly hedge accounting. On what the business actually earns, the market has not paid for the AI story at all.
The gap sits in a reconciliation table few people read, and even the data vendor's comparison with rivals uses the distorted figure.
The case against us: most AI demand is unsigned, Texas power prices for future delivery have softened, and data centres can build their own plants. A cut to the 2027 aim, rather than softer words, would prove us wrong.
What analysts say, and the evidence
The evidence under this - 4 claims
The vendor's separate top-level market-tier EV/EBITDA of 10.07x (S16) is close to the hand-computed Adjusted figure (C11), so not every vendor EBITDA field is equally distorted - the cohort-specific 21.1x figure appears to draw on a different, more GAAP-exposed quarter mix than the smoothed top-level ratio.
The call confirmed the higher range includes the Cogentrix plants closed on 30 October and the Meta PPA, which the old aim excluded, so some of the rise is acquired or contracted rather than a higher view of the old fleet; the 2027 interest expense in the guidance reconciliation is $1,430M, against $829M in the first nine months of 2026 (S24).
- 2026-11-06editedthe 6 November release initiated 2027 guidance above the $7.4-7.8B aim
- 2026-11-09editedThe Q3 call answered the open question: the 2027 range includes Cogentrix and the Meta Perry PPA, with no split given
Behind-the-meter capacity (about 3 GW by year-end 2026) remains a small fraction of the 12-15 GW of realistic new ERCOT load alone (C29), and cannot serve Vistra's 5-million-customer retail book or offer investment-grade counterparty risk transfer the way a PPA does (machine.threat.not_substituted).
This is the only publicly quantified large signed data-center commitment across Vistra's entire ERCOT/PJM footprint; there is no comparably sized signed PPA disclosed in ERCOT specifically.
What could go wrong, and how we would know
The risks
- Demand that routes around it. Data-centre owners building their own gas turbines or fuel cells never need to buy from Vistra's grid.
- Softer Texas prices. Battery buildout and a market redesign have lowered Texas power prices, and forecasts for 2026 demand growth were cut.
- Debt and buybacks together. Borrowing rose $2.6B in the six months to June, and the Cogentrix purchase and a $1B preferred redemption since add about $4.9B, while buybacks continue.
- A paused connection queue. A Texas audit has paused reviews of large new users, and it has already run longer than management expected.
The evidence under this - 8 claims
Behind-the-meter capacity (about 3 GW by year-end 2026) remains a small fraction of the 12-15 GW of realistic new ERCOT load alone (C29), and cannot serve Vistra's 5-million-customer retail book or offer investment-grade counterparty risk transfer the way a PPA does (machine.threat.not_substituted).
Management's own language - 'we expect to exhaust that no later than end of 2027' - is itself a target, not a commitment, and the same capital could instead go to additional debt paydown if the investment-grade ratings push takes priority, which management named as a live alternative use of cash.
- 2026-11-06editedthe 6 November release added $1.5B to the buyback authorization
- 2026-10-30editedThe acquisition closed on 30 October 2026, inside the mid-to-late 2026 window; the claim moves from forward to historical.
Management's own framing is self-interested - no company facing a queue audit would forecast material harm to its own timeline on the call announcing it - and the audit's scope and duration were still undefined at the time of the call.
What we still do not know
- How much Texas data-centre demand gets signed. There is no signed Texas deal of Meta's size yet.
- What Cogentrix's existing power contracts carry. They are private and not separately disclosed.
The evidence under this - 2 claims
This is the only publicly quantified large signed data-center commitment across Vistra's entire ERCOT/PJM footprint; there is no comparably sized signed PPA disclosed in ERCOT specifically.
- 2026-10-30editedThe acquisition closed on 30 October 2026, inside the mid-to-late 2026 window; the claim moves from forward to historical.
What would change our view
- Attractive→NeutralThe price rises within 20% of fair value
It needs:
- ○The share price rises to about $138, 80% of what we think the business is worth, so it is no longer cheap enough.about ≈$136.90 today, 1% above the last close
- Attractive→UnattractiveThe forecast the price pays for is cut
It needs:
- ○The 2027 profit forecast of $7.7B to $8.5B is cut, not just talked down, because realised power prices fell.checked at the next results, due by 26 Feb
- Attractive→UnattractiveThe price runs 20% above fair value
It needs:
- ○The share price rises to about $207, 20% above what we think the business is worth.about ≈$181.69 today, 33% above the last close
We would also be wrong if
- The Cogentrix plants or the nuclear upgrades for Meta slip or are repriced.
- Debt grows faster than profit as buybacks and borrowing-funded growth continue together.
- Building power on site becomes the normal way to supply a new data centre, bypassing the grid.
What is coming
- 15 December 2026 Series B preferred first reset date: Whether the second $1.0B of preferred is redeemed, completing the swap of preferred for junior notes.
Every dated event ahead
- 15 December 2026 Series B preferred first reset date: Whether the second $1.0B of preferred is redeemed, completing the swap of preferred for junior notes.
- 26 February 2027 Q4 2026 results and 2027 guidance update: Whether the 2027 Adjusted EBITDA range of $7.7-8.5B holds, how much of it is Cogentrix, Batch Zero timing, and signed load beyond Meta and the 207 MW New Era deal.
Since we wrote this
- 13 NovPrice moveStill Attractive: the rise left the price at 78% of valueThe shares rose about 8% on no news. Counting debt, which does not move with the price, they are about 78% of our $173 value, still inside the cushion; the view turns Neutral near $138.
- 9 NovEarningsThe call said the 2027 range includes Cogentrix and MetaThe raise over the old aim is partly bought and contracted plants, not a higher view of the old fleet. Leverage is 2.8 times and the range was not cut, so Attractive and Medium hold.
- 9 NovThesis reviewStill Attractive: priced at 71% of a value near $177Re-valued in three stages at an 8% return, the method now drawn on every thesis: the price sits more than 20% under the base case.
- 6 NovEarningsAttractive: forecasts raised, shares at 8.6 times profitQ3 underlying profit rose 50% and the first 2027 range sat above the old aim, while the shares fell 30% in a month; conviction stays Medium until the call is read.
- 31 OctThesis reviewThe page re-said from NeutralThe 12 October move to Neutral had left the verdict and the routes arguing from Attractive.
- 30 OctNewsCogentrix acquisition closed on scheduleThe 5,500 MW gas purchase closed on 30 October inside the expected window; the view holds pending the early-November results.
- 15 OctNewsVistra redeems its $1.0B 8.0% Series A preferred stock, as the September notes signalled. View heldVistra redeems all $1.0B of its 8.0% Series A preferred stock on the 15 October reset date.
- 12 OctThesis reviewView moved from Attractive to NeutralThe price discount that made up most of the case has nearly gone and the business has not changed.
- 6 OctNewsEnergy Department offers Vistra a conditional loan of up to $4.2B for nuclear plants. View heldEnergy Department said to offer Vistra a conditional loan commitment of up to $4.2B for nuclear uprates.
- 5 OctThesis reviewThe long case now sits under the verdictAfter the five-name pilot, the owner asked for the long case on every thesis.
- 4 OctThesis reviewOwner's checklist: revenue streams, cost of growth, management, catalystsThe owner's checklist found the method silent on who runs the company, what growth costs, where the money comes from and what is dated ahead; this pass fills those four blocks from the proxy and filings.
- 15 SepThesis reviewTwo open questions turned out stale. View heldThe grid rule filings had been made months earlier, and the Texas queue audit ran past its expected two months. Nothing was read to a primary source, so no claim moved.
- 24 AugThesis writtenAttractive, medium convictionTwo weeks after a quarter whose reported earnings missed by 52.8% on paper hedge losses while underlying profit grew over 30%.
Also checked 22 times with nothing new to act on.
Sources 25
Sources - 25
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S01primarycompany filing2026-02-27fleet composition, segments, nuclear PTC, Meta PPA, Cogentrix and Lotus transaction terms, competition section, risk factorsbacksC01C02C07C08C18C19C20C21C22C26C27C42C43C48C49
- S02primarycompany filing2026-08-10balance sheet, debt schedule, share count, EPS, liquidity as of 2026-06-30backsC16C17C43C53
- S03primarycompany filing2026-08-07Adjusted EBITDA by segment, guidance ranges, hedging loss detail, buyback and liquidity figures, Cogentrix FERC approval, Helix announcementbacksC03C04C05C06C13C14C15C21C24C39
- S04primarytranscript2026-08-07Vistra Q2 2026 earnings call transcriptmanagement's ERCOT/PJM load-growth framing, Batch Zero audit color, Helix detail, capital allocation and leverage targets, RTC+B/battery commentary
we put out a load forecast a little over two years ago in May of 2024 with an expectation of 115 to 120 gigawatts of load in ERCOT in 2030 ... we think is somewhere in the 12 GW to 15 GW by 2030. When you have numbers overstated by more than 20x, it causes problems for policy makers.
backsC13C14C17C20C23C24C25C26C27C29C32C34C37C38 - S05primarycompany filing2026-02-24confirmation of integrated generation-plus-retail model as a cohort peerbacksC41
- S06primarycompany filing2026-02-26confirmation of pure-play PJM merchant generation as a cohort peerbacksC41
- S07primarycompany filing2026-02-24confirmation of generation-plus-retail model as a cohort peer, nuclear-heavy asset mixbacksC41
- S08secondarynews2026-08-24Utility Dive - EIA cuts 2026 power generation forecast, ERCOT/PJM demand growth EIA (via Utility Dive)independent EIA-sourced ERCOT and PJM electricity demand growth rates
the EIA revised its forecast of ERCOT growth rates, shrinking its projections from 6% to 5% growth in 2025 and 15.7% to 9.6% growth in 2026
backsC30 - S09secondarynews2026-08-24Latitude Media / RTO Insider / Utility Dive - ERCOT large-load interconnection queue ERCOT (via trade press)independent size of the ERCOT large-load speculative queue
ERCOT's large load interconnection queue sits at 410 GW, about 87% of which are data centers
backsC28 - S10secondarynews2026-08-24Utility Dive - ERCOT 2030 load growth planning projections ERCOT (via trade press)independent ERCOT-sourced systemwide 2030 peak demand estimate
ERCOT anticipates about 152 GW of new load on its system by 2030
backsC31 - S11secondaryindustry data2026Cleanview - Bypassing the Grid: How Data Center Developers Are Building Their Own Power Plants Cleanviewindependent tracking of behind-the-meter/off-grid data-center generation capacity
approximately 2 GW of behind-the-meter generation capacity is online across four projects, with the vast majority belonging to xAI
backsC35 - S12secondarygovernment data2025-12-05ERCOT RTC+B go-live market notices and trade coverage (GridBeyond, ercot.com) ERCOTconfirmation and date of ERCOT's scarcity-pricing market redesign
Under the RTC+B regime, the ORDC adder to the energy prices is removed, and replaced by adders to real time ancillary service prices, known as ASDC
backsC33 - S13secondarygovernment data2026EIA - Plant Vogtle Unit 4 begins commercial operation / US nuclear construction status EIAindependent confirmation that no new large nuclear capacity is being added, supporting the scarcity of Vistra's existing licensed fleet
No nuclear reactors are under construction now in the United States.
backsC36 - S14secondarynews2026-08-23The Motley Fool - Vistra Is the Quietest Big Winner of the AI Power Boomevidence of the consensus AI-power framing of the stockread, but no claim rests on it
- S15secondarynews2026-08-07Reuters - Vistra quarterly profit slips on hedging losses despite strong power demandindependent read on the Q2 2026 GAAP miss and its hedging-loss driverread, but no claim rests on it
- S16secondaryvendor data2026-08-24Portfolio-management research pack (market tier: price, insiders, analysts, congressional, peer valuation)content/VST/research-pack.jsonprice/volatility snapshot, insider Form 4 aggregation, analyst consensus, congressional disclosures, cohort valuation multiplesbacksC39C40
- S17primarycompany filing2026-08-24data/reference/competitors.json - VST filed-competitor-class cohort reviewdata/reference/competitors.jsonthe research-tier cohort build (NRG, TLN, CEG) and each peer's confirming quotebacksC41
- S18primarycompany filing2026-02-26the original 2026 guidance ranges and the agreements announced alongside full-year 2025 resultsread, but no claim rests on it
- S19primarycompany filing2026-05-07Q1 2026 buyback pace and Adjusted EBITDA progress, corroborating the full-year buyback and guidance trendbacksC13C14
- S20primarycompany filing2026-03-18executive officer and director biographies, beneficial ownership as of March 3, 2026, the 2025 annual incentive scorecard and payout, the PSU design and the 2023 PSU resultbacksC45C46C47
- S21primarycompany filing2026-09-24$850M 7.000% Series A and $650M 7.250% Series B junior subordinated notes due 2057, issued by Vistra Operations and guaranteed by VistrabacksC49
- S22primarycompany filing2026-09-10use of proceeds: general corporate purposes including redeeming some or all of the Series A and Series B preferred stock at or after their October and December 2026 reset datesbacksC49
- S23primarycompany release2026-10-30
Vistra Corp. today announced it has completed the acquisition of Cogentrix Energy, consisting of 10 modern natural gas generation facilities totaling approximately 5,500 MW of capacity, from funds managed by Quantum Capital Group. The transaction closing follows receipt of all required regulatory approvals.
backsC21 - S24primarycompany filing2026-11-06Q3 Ongoing Operations Adjusted EBITDA by segment, narrowed 2026 and first 2027 guidance, buyback authorization, liquidity, Cogentrix cash and debt repaid, New Era agreementbacksC04C05C14C15C50C51C53
- S25primarytranscript2026-11-06what the 2027 range includes, leverage and ratings, Cogentrix economics, Batch Zero audit timing, the per-share free cash flow path, ERCOT forwards, buyback pace
We are introducing guidance ranges for 2027 adjusted EBITDA of $7.7 billion to $8.5 billion and 2027 adjusted free cash flow before growth of $4.3 billion to $5.1 billion, including the expected contribution from the Cogentrix acquisition and the power purchase agreement with Meta for the operating capacity at our Perry nuclear plant.
backsC05C52