Combined average of Moat (AI Resilience), Growth, and Valuation scores.
Moat Score
Talen's moat is a single NRC-licensed nuclear station — 90% of 2.5 GW Susquehanna, licensed through 2042 and 2044 — plus a 17-year, 1,920 MW front-of-the-meter nuclear PPA with Amazon that ramps no later than 2032, sitting on a 15.7 GW PJM-heavy merchant fleet. That is real plant-and-contract lock-in, but it is smaller and more concentrated than Constellation's ~22 GW nuclear book or Vistra's 6.4 GW nuclear plus Texas retail stack. Talen Energy Supply filed Chapter 11 in May 2022 after gas-price spikes squeezed liquidity and emerged May 17, 2023 after a plan that cut TES debt by about $2.7 billion; the current equity is the post-emergence story, not a 50-year utility franchise.
Talen is a PJM merchant generator with one nuclear station and one large hyperscaler offtaker, not a diversified nuclear platform. The durable pieces are the NRC license and the Amazon PPA; everything else is scale, mix, and balance-sheet history:
- Susquehanna NRC License — Real, Narrow: Talen operates and owns 90% of the 2.5 GW Susquehanna BWR station (owned capacity 2,245 MW). The two unit licenses expire in 2042 and 2044. New U.S. reactors still take a decade-plus of NRC process and billions of dollars, so the license is a genuine barrier — but it is one plant. CEG runs ~21 nuclear plants / ~22 GW; Vistra runs 4 plants / 6 reactors / 6.4 GW. A multi-month Susquehanna outage is a company-level event at Talen in a way it is not at those two.
- Amazon PPA: 1,920 MW Through 2042, Single Counterparty: The June 11, 2025 IR/8-K PPA commits 1,920 MW of carbon-free nuclear power to Amazon through 2042, ramping to full volume no later than 2032, with Talen as retail electric generation supplier and PPL on transmission after a spring-2026 front-of-the-meter cutover (concurrent with a Susquehanna refueling outage). The 2024 Cumulus campus sale (gross $650 million to AWS) created the original co-located relationship; FERC/ISA issues on the behind-the-meter path are why the structure is now grid-connected. The contract is the load-bearing embedding pillar. It is also Amazon-only — CEG's book spans Microsoft, Meta, Alphabet plus incremental corporates; Vistra has Amazon and Meta plus ~5 million Texas retail customers.
- Merchant PJM Fleet, Post-Restructuring Balance Sheet: Freedom and Guernsey (~2.8 GW CCGTs) closed November 25, 2025; Cornerstone (Lawrenceburg, Waterford, Darby, ~2.6 GW) closed June 15, 2026 for $3.45 billion. Owned capacity is 15,670 MW. July 31, 2026 net debt was $9.17 billion against a $2.125 billion FY26 adj. EBITDA midpoint (~4.3×, with only a partial Cornerstone year in the denominator) versus a <3.5× target. S&P/Fitch rate the unsecured notes BB-; Moody's Ba3. The 2022 bankruptcy is closed, but it is why this name does not get the same balance-sheet or franchise credit as CEG or Vistra.
Moat Verdict
Talen is an AI-power beneficiary through the Amazon PPA, not an AI-vulnerable software franchise. The load-bearing pillars are regulatoryLockIn (NRC licenses) and transactionEmbedding (the Amazon contract). Four software-style pillars are N/A on purpose. The 2022–23 bankruptcy, single-plant nuclear book, and Amazon concentration are why this moat sits below CEG and VST: same industry, thinner and more concentrated version of the same two pillars. Primary risk is Amazon ramp delay or a Susquehanna outage against a levered post-acquisition balance sheet.
65.0 resilient · 65.0 vulnerable · 80/20 = 65.0 · = 65
Open a moat to read its note.
Talen is a power generator and wholesale/retail supplier; there is no user-facing product interface that creates switching costs.
Susquehanna fuel-cycle, outage, and NRC-compliance practice is plant-specific operating knowledge. It is real and not software-replicable, but it is one two-unit BWR station — not CEG's 21-plant nuclear operating system.
Talen does not control access to a unique public data source; this pillar does not apply to a merchant generator.
NRC-licensed operators at Susquehanna (~900 site employees in the June 2025 Amazon release) are scarce. Intact, not strong: Talen staffs two units, not CEG's 21 plants or Vistra's six reactors, so the talent barrier does not constrain industry capacity the way those two do.
Talen now bundles generation with Pennsylvania retail supply to Amazon and a ~4 GW powered-land / new-capacity pipeline. The 2024 Cumulus campus sale removed the owned data-center bundle; what remains is replicable by other PJM generators with spare interconnection and land.
Decades of Susquehanna operating and fuel-cycle data help run that plant. That is plant telemetry, not a unique non-replicable dataset. Intact, not strong — the same call CEG and VST get for ops history, without their multi-plant or retail datasets.
NRC operating licenses on Susquehanna Units 1 and 2 run to 2042 and 2044; interconnection and PJM qualified-scheduling status sit on top. New nuclear entry is still a decade-plus, multi-billion-dollar process, but every US reactor owner holds the same class of licence, and Constellation, Vistra and PSEG sell nuclear output into the same hyperscaler contracts. A barrier every nuclear incumbent clears is intact, the rule applied to bank charters and exchange designations. Re-rated from strong to intact.
No network effects in power generation. Electrons on PJM are a commodity; additional offtakers do not raise the value of existing ones.
The 1,920 MW Amazon PPA through 2042 is a real 17-year bilateral embed, and Talen is the named retail supplier after the FTM cutover. Intact, not strong: one hyperscaler, one plant, and a ramp that is still mostly ahead (IR deck 480 MW contracted mix in 2028). CEG's multi-hyperscaler 15–20 year nuclear book and Vistra's Amazon+Meta plus 5 million retail accounts are the strong versions of this pillar.
Talen is not a system of record for any information function; this pillar does not apply to a generator.
Not a source of durability for this business: no structural unit-cost gap versus rivals that the other pillars do not already capture.
Sells a fungible product or capacity: buyers pay the market price, not a brand premium.
Moat in 5–10 years
Talen's moats hold, on the same footing as Constellation's and Vistra's. The Amazon PPA through 2042 is signed, the Susquehanna licences run to 2042 and 2044, and licensed operators stay scarce. Embedding would widen only with a second hyperscaler contract, and NRC licensing reform is the one thing that could loosen the licence barrier, but neither has moved far enough to call a direction.
No moat trending · 65 holds
Combined average of Moat (AI Resilience), Growth, and Valuation scores.
Moat Score
Talen's moat is a single NRC-licensed nuclear station — 90% of 2.5 GW Susquehanna, licensed through 2042 and 2044 — plus a 17-year, 1,920 MW front-of-the-meter nuclear PPA with Amazon that ramps no later than 2032, sitting on a 15.7 GW PJM-heavy merchant fleet. That is real plant-and-contract lock-in, but it is smaller and more concentrated than Constellation's ~22 GW nuclear book or Vistra's 6.4 GW nuclear plus Texas retail stack. Talen Energy Supply filed Chapter 11 in May 2022 after gas-price spikes squeezed liquidity and emerged May 17, 2023 after a plan that cut TES debt by about $2.7 billion; the current equity is the post-emergence story, not a 50-year utility franchise.
Growth Score
Q2 2026 (reported Aug 5) printed adj. EBITDA of $374 million (vs $90 million) and adj. FCF of $212 million (vs −$78 million) on $747 million operating revenue; GAAP net loss attributable to stockholders was $92 million (−$2.00/share) on 14.1 TWh of generation. Q1 2026 was stronger: $1.129 billion revenue, $63 million GAAP net income ($1.38 basic EPS), $473 million adj. EBITDA, $350 million adj. FCF, 15.6 TWh. H1 adj. EBITDA is $847 million and adj. FCF $562 million. Management raised FY26 adj. EBITDA to $2,025–$2,225 million and adj. FCF to $1,200–$1,350 million after Cornerstone closed, and the Aug 5 IR deck lifted the 2027/2028 adj. FCF/share outlook to ~$34 / ~$40 on a flat 47.9 million share count (~$37 / ~$48 with assumed buybacks). The 2025–26 EBITDA step-up is mostly Freedom, Guernsey, and Cornerstone, not a perpetual 40% organic rate. Talen Energy Supply's May 2022 Chapter 11 and May 17, 2023 emergence is the starting balance sheet for this growth, not a footnote.
Valuation Score
At $314.46 (Friday August 21, 2026 close) — near the $301.45 52-week low and ~30% below the $451.28 52-week high — TLN sits between bear ($200) and base ($390), about 19% below the 12–24 month base. Yahoo market cap is ~$15.1B on 47.9 million shares. July 31 net debt of $9.17B implies EV ~$24.2B, or ~11× the $2.125B FY26 adj. EBITDA midpoint and ~12× the $1.275B FY26 adj. FCF midpoint. The Aug 5 IR deck framed 2026/27/28 adj. FCF yields of ~8% / ~11% / ~14% at $334.10 (July 31); at $314 those yields are higher. Yahoo's 1-year target average is $465.50. Base is 12–24 month fair value if the $34 2027 FCF/share outlook prints, not the $451 cycle high.
One Nuclear Plant, One Hyperscaler
Talen is a PJM merchant generator with one nuclear station and one large hyperscaler offtaker, not a diversified nuclear platform. The durable pieces are the NRC license and the Amazon PPA; everything else is scale, mix, and balance-sheet history:
- Susquehanna NRC License — Real, Narrow: Talen operates and owns 90% of the 2.5 GW Susquehanna BWR station (owned capacity 2,245 MW). The two unit licenses expire in 2042 and 2044. New U.S. reactors still take a decade-plus of NRC process and billions of dollars, so the license is a genuine barrier — but it is one plant. CEG runs ~21 nuclear plants / ~22 GW; Vistra runs 4 plants / 6 reactors / 6.4 GW. A multi-month Susquehanna outage is a company-level event at Talen in a way it is not at those two.
- Amazon PPA: 1,920 MW Through 2042, Single Counterparty: The June 11, 2025 IR/8-K PPA commits 1,920 MW of carbon-free nuclear power to Amazon through 2042, ramping to full volume no later than 2032, with Talen as retail electric generation supplier and PPL on transmission after a spring-2026 front-of-the-meter cutover (concurrent with a Susquehanna refueling outage). The 2024 Cumulus campus sale (gross $650 million to AWS) created the original co-located relationship; FERC/ISA issues on the behind-the-meter path are why the structure is now grid-connected. The contract is the load-bearing embedding pillar. It is also Amazon-only — CEG's book spans Microsoft, Meta, Alphabet plus incremental corporates; Vistra has Amazon and Meta plus ~5 million Texas retail customers.
- Merchant PJM Fleet, Post-Restructuring Balance Sheet: Freedom and Guernsey (~2.8 GW CCGTs) closed November 25, 2025; Cornerstone (Lawrenceburg, Waterford, Darby, ~2.6 GW) closed June 15, 2026 for $3.45 billion. Owned capacity is 15,670 MW. July 31, 2026 net debt was $9.17 billion against a $2.125 billion FY26 adj. EBITDA midpoint (~4.3×, with only a partial Cornerstone year in the denominator) versus a <3.5× target. S&P/Fitch rate the unsecured notes BB-; Moody's Ba3. The 2022 bankruptcy is closed, but it is why this name does not get the same balance-sheet or franchise credit as CEG or Vistra.
Moat Verdict
Talen is an AI-power beneficiary through the Amazon PPA, not an AI-vulnerable software franchise. The load-bearing pillars are regulatoryLockIn (NRC licenses) and transactionEmbedding (the Amazon contract). Four software-style pillars are N/A on purpose. The 2022–23 bankruptcy, single-plant nuclear book, and Amazon concentration are why this moat sits below CEG and VST: same industry, thinner and more concentrated version of the same two pillars. Primary risk is Amazon ramp delay or a Susquehanna outage against a levered post-acquisition balance sheet.
65.0 resilient · 65.0 vulnerable · 80/20 = 65.0 · = 65
Open a moat to read its note.
Talen is a power generator and wholesale/retail supplier; there is no user-facing product interface that creates switching costs.
Susquehanna fuel-cycle, outage, and NRC-compliance practice is plant-specific operating knowledge. It is real and not software-replicable, but it is one two-unit BWR station — not CEG's 21-plant nuclear operating system.
Talen does not control access to a unique public data source; this pillar does not apply to a merchant generator.
NRC-licensed operators at Susquehanna (~900 site employees in the June 2025 Amazon release) are scarce. Intact, not strong: Talen staffs two units, not CEG's 21 plants or Vistra's six reactors, so the talent barrier does not constrain industry capacity the way those two do.
Talen now bundles generation with Pennsylvania retail supply to Amazon and a ~4 GW powered-land / new-capacity pipeline. The 2024 Cumulus campus sale removed the owned data-center bundle; what remains is replicable by other PJM generators with spare interconnection and land.
Decades of Susquehanna operating and fuel-cycle data help run that plant. That is plant telemetry, not a unique non-replicable dataset. Intact, not strong — the same call CEG and VST get for ops history, without their multi-plant or retail datasets.
NRC operating licenses on Susquehanna Units 1 and 2 run to 2042 and 2044; interconnection and PJM qualified-scheduling status sit on top. New nuclear entry is still a decade-plus, multi-billion-dollar process, but every US reactor owner holds the same class of licence, and Constellation, Vistra and PSEG sell nuclear output into the same hyperscaler contracts. A barrier every nuclear incumbent clears is intact, the rule applied to bank charters and exchange designations. Re-rated from strong to intact.
No network effects in power generation. Electrons on PJM are a commodity; additional offtakers do not raise the value of existing ones.
The 1,920 MW Amazon PPA through 2042 is a real 17-year bilateral embed, and Talen is the named retail supplier after the FTM cutover. Intact, not strong: one hyperscaler, one plant, and a ramp that is still mostly ahead (IR deck 480 MW contracted mix in 2028). CEG's multi-hyperscaler 15–20 year nuclear book and Vistra's Amazon+Meta plus 5 million retail accounts are the strong versions of this pillar.
Talen is not a system of record for any information function; this pillar does not apply to a generator.
Not a source of durability for this business: no structural unit-cost gap versus rivals that the other pillars do not already capture.
Sells a fungible product or capacity: buyers pay the market price, not a brand premium.
Moat in 5–10 years
Talen's moats hold, on the same footing as Constellation's and Vistra's. The Amazon PPA through 2042 is signed, the Susquehanna licences run to 2042 and 2044, and licensed operators stay scarce. Embedding would widen only with a second hyperscaler contract, and NRC licensing reform is the one thing that could loosen the licence barrier, but neither has moved far enough to call a direction.
No moat trending · 65 holds
Growth Analysis
Growth Drivers
Key Risk
If Amazon holds deliveries at the 480 MW 2028 minimum into 2029–30, or a Susquehanna extended outage takes the 2.2 GW nuclear station offline for a quarter while July 31 net debt is $9.17B (~4.3× the $2.125B FY26 EBITDA midpoint), the 2027–28 adj. FCF/share outlook of ~$34 / ~$40 would miss and the multiple would compress. Falsifier: contracted Amazon MW still at the 480 MW 2028 step into calendar 2029, or FY26 adj. EBITDA below the $2,025M guide floor.
Score Derivation
73.6 base + 2.7 trajectory + 4 margin − 10 risk = 70
Base ~74 on 10.5% midpoint of 8–13% + 2.7 trajectory (Amazon PPA ramp and acquired-fleet contribution accelerating; PJM capacity/sparks already in the 2027–28 outlook, treated stable) + 4 expanding adj. EBITDA/FCF − 10 high keyRisk (Amazon concentration + single nuclear plant + ~4.3× July 31 net leverage) = 70.
Growth Drivers (3-Year Horizon)
Amazon PPA: 1,920 MW through 2042, full volume no later than 2032; IR deck maps 480 MW of long-term contracted mix in 2028 and 1,920 MW by 2030. That is a MW ramp to a cap, not a perpetual growth rate.
M&A anniversary: Freedom/Guernsey (Nov 2025) and Cornerstone (June 2026) are inside the raised $2.025–$2.225B FY26 EBITDA guide. After 2027 they anniversary; they do not keep adding 100% to the CAGR.
PJM: >10 GW cleared at $325/MWd for 2028/29; 2027/28 West Hub sparks up ~50% on the deck vs mid-2025 marks. Capacity auctions reset; this is a priced tailwind, not a structural 15% forever.
Capital return: 550,000 shares / ~$200M bought in Q2 2026; $1.7B remaining on the SRP through December 2028; deck assumes ~70% of adj. FCF into buybacks. Per-share math is real only if FCF prints and the board keeps authorizing.
Price Scenarios (12–24 Months)
Price vs Targets
Dashed lines mark the 12–24 month bear, base and bull targets. Daily closes.
Valuation Multiples
| Trailing P/E (GAAP) | N/M |
| Price / FY26E Adj. FCF | ~12× |
| EV / FY26E Adj. EBITDA | ~11× |
| Price / 2027E Adj. FCF (flat share) | ~9× |
| Net debt / FY26E Adj. EBITDA | ~4.3× |
GAAP P/E is noise (TTM loss, derivative MTM, post-deal interest). Cash-flow multiples are the lens: ~11× FY26 adj. EBITDA and ~12× FY26 adj. FCF are not distressed for a contracted-plus-merchant PJM generator, and they are not CEG-like infrastructure multiples either. The $200 / $390 / $520 corridor is anchored to the IR FCF outlook and a merchant-with-PPA multiple, not to the $451 52-week high.
Approximate figures as of August 2026.
Where We Are vs Targets
Loading live price…
Amazon holds the PPA at the 480 MW 2028 minimum, a Susquehanna outage or PJM capacity reset cuts FCF, and ~4× net leverage leaves no room — the stock de-rates toward ~8× compressed cash flow.
- Amazon delays the ramp toward 1,920 MW; 2028–29 contracted MW stay near the 480 MW minimum shown on the IR deck, so the 2027–28 ~$34 / ~$40 FCF/share outlook misses
- An extended Susquehanna refueling or forced outage takes the 2,245 MW owned nuclear station offline for a quarter; one-plant concentration shows up in EBITDA and in the NDT/interest stack
- PJM 2028/29 $325/MWd capacity pricing does not repeat and/or PPL zonal basis stays wide; adj. EBITDA slips toward or below the $2.025B guide floor while $9.17B net debt stays, and the equity de-rates toward ~8× a compressed FCF print near $200
FY26 lands inside the raised $2.025–$2.225B adj. EBITDA / $1.20–$1.35B adj. FCF guide; 2027 adj. FCF/share prints near the ~$34 flat-share outlook; the multiple holds around ~11–12× FCF as Cornerstone annualizes.
- FY26 adj. EBITDA hits near the $2.125B midpoint and adj. FCF near $1.275B (~$26.60/share on 47.9M shares), validating the Aug 5 raise after Cornerstone
- Amazon FTM cutover completes with the 2026 Susquehanna outage and the PPA continues to ramp off the 480 MW 2028 step toward the 1,920 MW cap
- 2027 adj. FCF/share near $34 on a flat share count; at ~11.5× that FCF the stock is ~$390 — 12–24 month expected value, not a re-rating to the $451 52-week high
Amazon ramps faster than the 2032 backstop, one incremental data-center PPA converts from the ~4 GW pipeline, buybacks lift FCF/share toward the deck's ~$48 2028 print, and the multiple holds.
- Amazon reaches a larger share of the 1,920 MW cap before 2030, and the IR deck's 'existing PPA acceleration' lever shows up in 2028 FCF
- At least one incremental long-term data-center PPA is signed from the ~4 GW development pipeline (the 2030+ 4,000 MW mix on the deck is illustrative, not contracted)
- SRP execution at ~70% of adj. FCF plus the $1.7B remaining authorization lifts 2028 FCF/share toward ~$48; at ~11× that print the equity is ~$520