Thematic Exposure -- 5/10
Vistra rides a genuinely powerful theme -- AI/data-center-driven US power demand and ERCOT/PJM load
growth -- and is the largest competitive power generator in the US (~41 GW, ~43.7 GW including the
second-largest competitive nuclear fleet at 6,448 MW). But "largest competitive generator" is a
national-aggregation title, not a dominant share of any single power market. Across every meaningful
segment VST is a low-to-mid-double-digit share player with many >15%-class competitors and 100+
retail rivals. It fails the oligopoly hard gate -- no >30%-share segment, and competitive power is
not a ≤3-player ≥70% market. Big TAM does not compensate. Ceiling = 5/10.
Weight: 35%
AI / Data-Center Power Demand -- Strong Theme
Secular Tailwind -- Multi-Year Visibility
VST is the scaled ERCOT incumbent at the exact moment demand is inflecting. Management sees ERCOT
total load growth of ~30-40 GW by 2030 (10-15 GW data centers); US data-center power is projected
at 177-192 TWh (2024) rising to 325-580 TWh by 2030 (LBNL). PJM capacity prices are tightening, the
Meta nuclear PPA (~2,609 MW) is signed, and the theme is compounding well above GDP (>10% growth).
Co-Leader -- Not a Dominant Oligopolist
Oligopoly Gate: FAIL
VST holds ~13% of ERCOT (~19.6 GW of ~150+ GW), single-digit share of PJM (~180+ GW), and is #1
but fragmented in Texas retail against 100+ PUCT-licensed REPs. No single segment gives VST >30%
share, and no segment is a three-or-fewer-player >70% structure. Competitive power generation is a
contested market with many strong players -- NRG, Calpine, Constellation, Talen, PSEG -- not a true
oligopoly.
The Moat Is in the Assets, Not the Market Structure
Price-Taker in Commodity Merchant Power
VST owns ~41 GW of dispatchable + zero-carbon capacity that is extremely hard to replicate --
siting, interconnection queues, and the second-largest competitive nuclear fleet are multi-year,
multi-billion-dollar barriers. But at the molecule level power is a commodity: any new ERCOT entrant
(41 GW of Texas Energy Fund gas filed, plus merchant solar/storage) can sell the same MWh, PJM
capacity is contestable, and a retail customer can switch REP in well under 12 months. VST is a
price-taker whose edge is the installed, contracted, hard-to-permit fleet plus nuclear scarcity --
not a structural monopoly on the product.
Segment Revenue (FY2025, Daloopa)
Total reported Q1'26 revenue $5.64B after intersegment eliminations. Retail Adj EBITDA of $68M vs Generation Adj EBITDA of $1,426M in Q1'26 -- the business is ~95% generation-economics-driven, and retail revenue is largely intersegment.
Segment Table -- Share / TAM / Theme
| Segment | % of Rev | Market Share | Theme Growth |
|---|---|---|---|
| Competitive Gen -- Texas/ERCOT | ~53% | ~13% of ERCOT (~19.6 GW of ~150+ GW); leading competitive generator in Texas but NRG, Calpine, Luminant peers, munis/coops, solar/storage all present | >10% -- ERCOT load growth + AI/data-center |
| Competitive Gen -- East/PJM-MISO | ~40% | Single-digit % of PJM (~180+ GW); meaningful nuclear (Comanche Peak, Perry, Davis-Besse, Beaver Valley) but Constellation is the larger PJM/nuclear player | >10% -- PJM capacity prices rising; Meta PPA (2,609 MW) |
| Competitive Retail -- TXU + brands | ~5% EBITDA | #1 REP in Texas (~1.7M+ TXU; ~4.3M total) but 100+ PUCT-licensed REPs; highly fragmented | ~GDP-like volume; TXU premium pricing ~20% above market |
| West / Asset Closure / Sunset | ~2-5% | De minimis; legacy/coal wind-down | Declining (intentional) |
Competitors: ERCOT generation -- NRG, Calpine, Luminant peers, Engie, plus 41 GW of Texas Energy Fund gas filed and merchant solar/storage. PJM generation -- Constellation (larger nuclear), Talen, NRG, PSEG. Retail -- Reliant/NRG, Gexa, Rhythm, Constellation, Ambit, 100+ others.
Required Pre-Scoring Questions
| Question | Answer |
|---|---|
| >15%-share competitors per segment? | ERCOT: VST ~13% with NRG, Calpine, Luminant peers and a long fragmented tail. PJM: VST single-digit; Constellation/Talen/NRG/PSEG all sizable. Retail: 100+ REPs. Every segment fragmented or non-dominant. |
| Could a customer replace VST in 12 months? | Retail: yes, trivially (switch REP). Wholesale: a counterparty can source MWh elsewhere, though scarce firm/nuclear capacity is a partial exception. Largely yes -- moat is in the assets, not the product. |
| Price-maker or price-taker? | Price-taker in merchant wholesale (ERCOT/PJM clearing prices set realized margin; mgmt debates "mispricing in the forward curve"). TXU premium pricing is brand, not market power. |
Oligopoly Hard Gate
| Criterion | Result |
|---|---|
| VST share in ERCOT | ~13% |
| VST share in PJM | Single-digit |
| Any segment >30% share? | No |
| Key competitors | NRG, Calpine, Constellation, Talen, PSEG, 100+ REPs |
| Gate result | FAIL |
5/10 — Vistra has best-in-class
thematic exposure -- the largest US competitive generator sitting directly in the path of the
AI/data-center power-demand wave, with a scarce nuclear fleet and accelerating ERCOT/PJM load growth
(>10% theme growth), reinforced by the Meta nuclear PPA and the Cogentrix gas acquisition. That is a
real, durable asset-level moat built on un-replicable siting, interconnection, and nuclear scarcity.
But thematic strength is not market dominance: VST is a ~13% ERCOT generator, single-digit in PJM, a
#1-but-fragmented retailer against 100+ REPs, and a price-taker in commodity merchant power. With no
>30%-share segment and no ≤3-player oligopoly, the hard gate caps this at exactly 5/10 -- a
top-of-theme name whose competitive structure is too fragmented and price-taking to merit the leader
premium the rubric reserves for oligopolists.
Data sourced from Daloopa (company_id 27524).