CEG — Constellation Energy Corporation

As of 2026-09-13, the sell-side is firmly constructive: 15 Buy/Strong Buy ratings versus 6 Holds and zero Sells, with a consensus 12-month price target of $354.50 (high $441, low $296) against a $284.75 spot price - implying roughly 24% upside. The tape supports it partially: CEG was up 12.21% over the trailing three months and 2.18% over the trailing month as of 2026-09-13, and the August 6, 2026 Q2 print beat consensus ($2.55 adjusted EPS vs. ~$2.34) with FY2026 guidance raised to $11.50-$12.50 from $11.00-$12.00. The Meta-linked 20-year Clinton clean-energy PPA and continued data-center co-location chatter keep the structural AI-power narrative alive. What tempers it is the recent action: the stock fell 4.75% in the week into 2026-09-13 and sits 5.64% lower over six months, so sentiment is bullish on fundamentals but no longer momentum-driven.

The reported series is dominated by one event, not by organic trend. Revenue ran $6.1B-$7.2B per quarter through 2025, then jumped to $11.12B in Q1 2026 (+64% YoY) and $7.51B in Q2 2026 (+23% YoY) as Calpine consolidated - the deterministic score's own note flags a +104% sequential revenue move and a +131% jump in stockholders' equity between 2025-12-31 and 2026-03-31, which is balance-sheet evidence of the acquisition, not organic acceleration. Operating margin is genuinely volatile and seasonal: 20.1%, 22.4%, 18.1%, 6.6%, 15.6%, 21.4%, 2.7%, 20.8%, 7.7% across the last nine quarters - and the Q2 2026 compression to 7.7% (from 15.6% in Q2 2025) largely reflects the lower-margin gas fleet and purchase-accounting depreciation Calpine brought in, not a collapse in the nuclear business. Diluted EPS swung from $4.49 in Q1 2026 to $1.45 in Q2 2026 on a TTM base of $9.825. Free cash flow turned negative for three straight quarters (Q4 2025 -$181M, Q1 2026 -$850M, Q2 2026 -$118M) as quarterly capex stepped up to $1.25-$1.28B from under $1B - that is a real trend worth watching, and it is the single most credible bear signal in the dataset. Note the gross profit field for Q4 2025 ($15.7B on $5.46B of revenue) is internally inconsistent and makes the 94.9% TTM gross margin unreliable; don't build a thesis on it.

Mixed, leaning fair-to-rich depending on the lens. As of 2026-09-13, CEG trades at 27.67x TTM EPS of $9.825, which is actually below its own 5-year average TTM P/E of 29.94x, and at roughly 23.6x the $12.045 next-FY consensus EPS. But on asset and sales multiples it is clearly rich versus its own history: price/sales of 3.27 versus a 5-year average of 2.14, price/book of 3.15, and EV/EBITDA of 14.66 - high for an independent power producer, where 10-12x is more typical and premium nuclear names stretch to 13-14x. The quantitative score independently labels it "Rich vs. its own history" (reference score 29.7). Verdict: not a bubble, but the forward-earnings case only works if the $12.045 estimate is real and durable - and that estimate is heavily dependent on a Calpine fleet that only two quarters of reported results actually reflect.

As of 2026-09-13, the market is paying ~23.6x the $12.045 forward EPS estimate against a 5-year average TTM P/E of 29.94x - so the stock is being valued as if the Calpine-boosted earnings base is real but does not deserve the historical premium, which is a fair posture given 4% ROIC and $19.1B of long-term debt. The setup is asymmetric around the Q3 2026 print (late October/early November 2026): if CEG delivers inside or above the $11.50-$12.50 full-year guide, the multiple looks undemanding and there is room toward the $354 consensus target; if adjusted EPS drifts back toward the $9-$10 range as gas margins normalize and integration costs land, the same price is 28-31x earnings and the downside opens toward the $228.63 52-week low. Watch adjusted operating EPS specifically, not GAAP - the GAAP line moved from $4.49 to $1.45 in one quarter and will whipsaw on mark-to-market hedges.

Neutral and range-bound, not a timing signal in either direction. As of 2026-09-13, RSI(14) was 53.69 - dead center - with price 1.14% above the 20-day SMA and 5.65% above the 50-day SMA, but 3.56% below the 200-day SMA. That combination says the stock is in a short-term recovery inside a longer-term downtrend: up 12.21% over three months but still down 5.64% over six months and 4.75% in the week into 2026-09-13. The stock has spent the year carving out a wide $228.63-$412.70 range and is sitting in the middle of it. For entry timing, a pullback toward the 200-day average would be technically cleaner than buying into a mid-range, mid-RSI tape with a negative long-term moving average overhead; equally, a decisive reclaim of the 200-day would confirm the 3-month recovery. Neither has happened yet.

• Data-center and AI electricity demand is the core secular driver - hyperscalers want 24/7 carbon-free baseload, which is exactly what CEG's nuclear fleet sells, and long-dated PPAs (the 20-year Meta/Clinton deal, plus 920 MW of new nuclear PPAs signed in Q2 2026) convert that demand into contracted revenue visibility.
• Interest rates and leverage: long-term debt jumped from $7.25B (2025-12-31) to $19.11B (2026-06-30) post-Calpine, so both refinancing cost and the discount rate applied to a capital-intensive utility multiple are live headwinds, and the 0.625% dividend yield offers almost no income cushion versus a 10-year Treasury.
• PJM capacity market design and FERC regulation - CEG is actively litigating PJM's Interim Resource Adequacy Service proposal, and the outcome of capacity pricing and large-load co-location rules directly determines how much of the data-center upside CEG captures versus gives away.
• Power and natural gas price spreads: adding Calpine's gas fleet makes CEG's earnings more exposed to spark spreads and gas volatility than the old nuclear-only story implied, which cuts both ways depending on the commodity cycle.
• Sector rotation and utility rate sensitivity - CEG trades as a hybrid AI-growth/defensive utility, so it gets hit when growth stocks de-rate and when rate-sensitive defensives rotate out, as the -31% drawdown from the 52-week high demonstrates.

• Q3 2026 earnings, expected late October/early November 2026 - the first clean read on whether the raised $11.50-$12.50 FY2026 guidance is intact after full Calpine consolidation.
• Completion of the $715M RISEC (609 MW Rhode Island State Energy Center) acquisition from Shell, announced 2026-09-10 and expected to close in Q1 2027, management flags it as immediately accretive.
• Additional long-term nuclear PPAs and data-center co-location agreements following the 20-year Meta/Clinton deal and the 920 MW signed in Q2 2026.
• Crane Clean Energy Center restart milestones (FERC waivers and NRC fuel licensing already secured), which would add carbon-free megawatts into a tight PJM market.
• Further portfolio optimization - the $860M Brazos Valley gas plant sale to LS Power - plus continued execution on the $5B buyback program through 2027.
• PJM capacity auction results and FERC rulings on the Interim Resource Adequacy Service proposal.
• Federal nuclear production tax credit policy and any change to clean-energy attribute pricing.

• Balance-sheet transformation: long-term debt roughly tripled from $7.25B at 2025-12-31 to $19.11B at 2026-06-30, total liabilities reached $65.9B, and debt-to-equity of 0.772 understates the leverage if you adjust for the still-negative free cash flow.
• Chronically low returns on capital - ROIC of 4% against an ROE of 16% means the return gap is coming from leverage, not from business quality, and 4% ROIC is well below any reasonable cost of capital.
• Negative free cash flow in three consecutive quarters (Q4 2025 -$181M, Q1 2026 -$850M, Q2 2026 -$118M) with capex running $1.25-$1.28B per quarter - the dividend and buyback are being funded from somewhere other than internally generated cash.
• Regulatory and market-design risk at PJM/FERC - adverse rulings on capacity pricing or large-load co-location would directly cap the data-center revenue premium the stock is valued on.
• Earnings volatility and quality: diluted EPS swung $4.49 to $1.45 quarter-over-quarter and quarterly operating margins ranged from 2.7% to 22.4% over the last eight quarters, making the $12.045 forward estimate harder to underwrite than the headline P/E suggests.
• Valuation risk on non-earnings multiples: price/sales of 3.27 versus a 5-year average of 2.14 and EV/EBITDA of 14.66 leave little room for a miss, and the stock has already shown it can fall 31% from its high in under a year.

Not a buy at these levels with conviction, and not a sell either - this is a genuinely good asset base (largest carbon-free nuclear fleet in the U.S., real 15-20 year data-center contracts) carrying a balance sheet that just got much heavier and generating a 4% ROIC that does not yet justify a premium multiple. The deterministic score's "Weak Fundamentals" label is mostly an artifact of reading four quarters of data that straddle the largest acquisition in company history - the underlying business beat and raised guidance on 2026-08-06 and is guiding to $11.50-$12.50 against $9.825 TTM. But the counter-argument is equally concrete: three straight quarters of negative free cash flow, $19.1B of long-term debt, and a 14.66x EV/EBITDA mean you are paying a full price for a turnaround in returns that has not yet shown up in the cash flow statement. I would want either a lower price (closer to the low-to-mid $260s, nearer the 200-day average) or one or two more quarters proving Calpine accretion and FCF inflecting positive before treating this as an overweight.

I applied a 26x forward P/E - a modest premium to the ~18-22x typical utility/IPP range, justified by the contracted nuclear baseload and data-center PPA pipeline but discounted versus CEG's own 29.94x 5-year average TTM P/E to reflect 4% ROIC and the post-Calpine leverage - to the next-FY consensus EPS estimate of $12.045, which yields $313.17 (rounded to $313). Cross-checks agree on direction: at $313 the stock would trade at ~13.6x EV/EBITDA on ~$8.3B of TTM EBITDA, still above the 12-13x I would call fair for the asset mix, and it sits ~12% below the $354.50 sell-side consensus, consistent with my more conservative view on the durability of Calpine-boosted earnings and free cash flow.