1. Read a sudden M&A / IPO wave in a sector as a repricing signal — then find the cleanest survivor
The repeatable method
- When a sector sees a record-setting deal and a hot IPO inside the same week, treat it as the market repricing a resource in real time — the insiders (acquirers, underwriters) are telling you scarcity has arrived before the screens catch up.
- Don't chase the deal stock or the IPO pop. Ask what scarce thing the wave is bidding for (here: firm, dispatchable power), then screen the sector for the operator that already owns the most of it, cash-generative and contracted.
- Use the deal multiples as a valuation anchor: if a target is being bought at a premium, a comparable un-bid operator trading cheaper is the mispricing.
Here: the NEE-D ~$67B deal (biggest since Exxon-Mobil 1998) and the FRVO $7.7B IPO (popped 33%) within a week → the scarce thing is firm power → buy the diversified, cash-rich owner CEG rather than the pre-revenue IPO or the acquirer.
Watch for
- Two or more outsized capital events (mega-merger, IPO, take-private) clustering in one sector in a short window; a quality operator that hasn't been bid up yet trading below the deal multiples.
2. Rank power assets by capacity factor — firm baseload is the value driver in an AI-demand world
The repeatable method
- For any power-generation name, pull the capacity factor — the share of the year the plant actually produces at full output. It's the single number that separates "firm" (dispatchable) from "intermittent" generation.
- In a market where the marginal buyer is a 24/7 data center, weight the firm sources (nuclear ~95%, geothermal 70-95%, gas, hydro) and discount the intermittent ones (solar ~25%, wind ~35%) — regardless of headline "clean energy" labels.
- Prefer operators whose fleet skews to high-capacity-factor assets; that's where the contracted, premium-priced demand lands.
Here: solar ~25% / wind ~35% vs geothermal 70-95% / nuclear ~95% — so the thesis sits on CEG's ~21 GW nuclear (94.7% CF) + The Geysers geothermal, not on a wind/solar developer.
Watch for
- Disclosed capacity factors and fleet mix; data-center / hyperscaler siting near firm-power capacity as the demand confirmation.
3. Follow the tax-credit asymmetry — let policy tell you which energy sub-sector has the tailwind
The repeatable method
- Read the actual statute, not the press release: which technologies keep their subsidies and which lose them, and on what timeline.
- Tilt toward the sub-sector the policy protects — a deliberate asymmetry is a multi-year tailwind, and it's often telegraphed by who wrote the carve-out.
- Cross-check the policymaker's background for the bias (a former operator carving out their old sector is a signal, not a coincidence).
Here: the One Big Beautiful Bill gutted wind/solar credits (must be in service by end-2027) but kept the full ITC/PTC for geothermal/nuclear/hydro/storage through 2033 — and Energy Secretary Chris Wright (ex-Liberty Energy, a Fervo investor) carved geothermal out. HEATS Act + BLM exclusions reinforce it.
Watch for
- Statutory in-service deadlines and credit phase-downs by technology; permitting waivers (HEATS Act, BLM categorical exclusions) and land auctions as on-the-ground confirmation.
4. Treat long-dated hyperscaler PPAs as contracted-revenue de-risking — price the annuity, not the merchant tape
The repeatable method
- When a generator signs multi-year power-purchase agreements with creditworthy hyperscalers, re-rate the locked-in revenue toward a regulated-annuity multiple rather than a volatile merchant-power multiple.
- Quantify the committed demand (MW under contract × contract length) versus the fleet, and check the counterparties' credit — that's the floor under the cash flows.
- Discount the still-merchant portion separately; the mispricing is usually the market valuing the whole company at merchant risk while a growing slice is contracted.
Here: CEG's 20-year PPAs — MSFT (Crane/TMI, 835 MW), META (Clinton, 1,121 MW), CyrusOne (Freestone, 380 MW) = >2,700 MW of committed long-duration demand — the contracted backbone of the "Wall Street hasn't repriced it" claim.
Watch for
- New PPA announcements (MW, term, counterparty); the contracted-vs-merchant revenue split widening; restart/online dates that bring contracted volumes live.
5. Size the upside to a named regulatory catalyst — buy up to a price, not at any price
The repeatable method
- When the bull case depends on a specific, dated regulatory event, isolate it: identify every approval needed and who opposes it.
- Set a buy-up-to price that leaves margin for the catalyst slipping — the contracted base supports a floor, the catalyst supplies the upside.
- Re-rate as the approvals clear (or as a slip pushes the contracted revenue out a year).
Here: "buy CEG up to $320" (~$289 now) — the upside hinges on the Crane restart clearing the NRC (safety), FERC + PJM (interconnection transfer from Eddystone, opposed by PJM's market monitor) by 2027; a slip discounts the Microsoft revenue.
Watch for
- NRC/FERC/PJM docket milestones on the Crane restart; the buy-up-to level versus spot; declining-peak price structure as the entry discipline.