Momentum and Socrates Arrive Early, Then the Stock Sells With the AI-Power Trade: At $67.86 the Bear Case Is the Price

Results vs. Consensus

Williams reported after the close on Monday, Nov. 2, and held its call at 9:30 a.m. ET the next morning, which was election day. After a second quarter that carried a joint venture and an acquisition announcement, this was a quarter of delivery rather than news. The results were on consensus, the guidance range held, and the operating milestones set in August arrived on or ahead of time. The news that moved the stock came from the election results that evening.

Q3 2026 Scorecard

The EPS panels split by a cent. Three carried $0.54 into the print and recorded an in-line result; two carried $0.55 and recorded a 1.8% miss. Previews from mid-October sat at $0.56, so the estimate drifted down into the print and the company met the revised figure. The revenue spread is again a modelling difference rather than a disagreement about the business, and for once it hardly matters: the derivative mark inside reported revenue was a $2M gain, so the $3,223M beat every published figure. No total adjusted EBITDA consensus was published. Unlike the second quarter, estimates were available for all five segments, and their sum is about $2,058M; the largest segment's estimate was published only in rounded form at $1.03B, so we label the comparison approximate.

Income Statement, Year Over Year

Sequentially, vs. 2Q 2026

Sequential growth of $118M is the cleanest read on the second half so far. Transmission, Power & Gulf added $71M as Socrates South produced its first full quarter of power and storage rates kept rising, and West added $42M, of which $40M is Momentum's first 28 days. The Northeast was flat on a record base. GAAP EPS fell sequentially because the second quarter carried $282M of net pre-tax items that adjusted results exclude: the $126M Brazos Permian II gain, $142M of unrealized derivative gains ($120M in Sequent and $22M in Other), the $12M South Mansfield gain and $4M of linefill, less $2M of intangible amortization. Without the Brazos gain alone, second-quarter GAAP net income would still have been above the third quarter's $699M, so the gains and derivative marks together explain the decline. Leverage rose 0.13x because Momentum's debt landed on the balance sheet with less than one month of its earnings in the trailing twelve months.

Revenue

For the first time in our coverage, reported revenue is a usable measure of the quarter. The mark-to-market derivative line that swung reported revenue by $359M in the first quarter and $94M in the second contributed $2M here, so the 10.3% reported growth and the 12.3% underlying growth are within two points of each other. The composition is what matters. Service revenues of $2,319M grew $198M, and the release names four sources: new power generation revenue, Momentum, projects placed in service and higher storage revenue. The quarterly filing sizes the first of these at a $53M year-over-year increase in power revenue in Transmission, Power & Gulf, which includes reimbursable fuel. Momentum contributed $113M of revenue in its first 28 days, and Gulf Coast storage revenue rose $21M on higher rates. Product sales of $845M grew 20.5% on higher NGL marketing volumes and Momentum, and the matching cost line grew faster.

Margins and operating leverage

Operating income rose only 2.4% on revenue up 10.3%, and the reason is entirely on the expense side. General and administrative expense rose $107M to $275M, and the release attributes $63M of the period's higher operating and administrative expense to acquisition and joint-venture formation costs that management excludes from adjusted results. Depreciation rose 9.9% to $620M on assets placed in service, Momentum, and the absence of last year's rate-case depreciation adjustments at Transco. Operations and maintenance rose 9.6%, with Momentum operating costs and operating taxes named. A $83M gain on the Permian gathering sale sits inside total costs and expenses and flatters operating income by that amount. Taking the gain out and adding the transaction costs back, operating income was roughly flat. That is not a deterioration: depreciation on new assets arrives before their full earnings do, and the third quarter carried one month of Momentum against its full integration cost.

EPS and below the line

Interest expense of $385M rose 3.5% year over year and 3.8% sequentially. The September note issue added $2.75B at coupons between 5.000% and 6.400%, and Momentum brought $869M of retained secured debt, so the fourth quarter carries the full cost of both. The unchanged guidance bridge of $1,535M implies $403M in the fourth quarter, which looks achievable rather than conservative. Other income rose $29M on equity AFUDC from higher eligible capital spending at Transco and Line 200, a line that grows with the construction program and fades as projects enter service. The line that matters most is noncontrolling interests, which nearly doubled to $67M from $36M and rose $18M sequentially. The quarterly filing attributes the increase primarily to higher results at Pantheon, the 51%-owned entity that holds the five power projects. This is the first full quarter in which the joint-venture partner's share of power earnings is visible, and it is running ahead of the guidance bridge.

Segment Performance

Momentum memo rows deduct the $40M of Modified EBITDA the acquired operations contributed from Sept. 3 to Sept. 30, as disclosed in the quarterly filing. The acquisition costs are excluded from adjusted EBITDA and are not part of that $40M.

The variance against estimates is small and concentrated. Transmission, Power & Gulf landed on its estimate, Northeast beat by 0.8%, and the two small segments missed by $7.5M together. West accounts for most of the shortfall, missing by $16M or 3.8% with Momentum's first month already inside the number. That is the second consecutive quarter in which West has been the weakest segment against expectations, and the memo row explains why: the legacy West business shrank 1.6% year over year once the acquisition is taken out. The rest of the portfolio grew 7% to 9% in its three infrastructure segments, and the total grew 6.2% with the marketing segment contributing almost nothing, the same shape as the second quarter.

Transmission, Power & Gulf

The largest segment grew 8.8% to $1,030M, an acceleration from 6.2% in the second quarter, and power is the reason. The segment's other fee revenues rose to $59M from $6M a year ago, a $53M increase that matches the power revenue increase the quarterly filing attributes to Socrates South and reimbursable fuel. Gulf Coast storage revenue rose $21M on higher storage rates, and the Whale and Overthrust Westbound expansions added $7M each. Against that, segment operating and administrative costs rose $63M to $353M, which includes the $30M of joint-venture formation costs the company excludes from adjusted EBITDA, fuel that is passed through to the power customer, and higher employee costs and corporate allocations. Regulated interstate transportation and storage revenue was flat at $930M, because the Transco rate increase has now annualized.

"growth in this segment was led by the initial contributions from our Socrates project, and continued growth from our natural gas storage businesses and pipeline expansions."
— John Porter, Chief Financial Officer

Assessment: this is the first quarter in which the power franchise shows up in a reported number, even if only inside a line called other fee revenues. Socrates South ran the full quarter and Socrates North adds the remaining ~356 MW this month, so the fourth quarter should show the segment's growth stepping up again. The flat regulated revenue line is the reminder that Transco itself grows in steps, at in-service dates, and the next large step is Southeast Supply Enhancement in 2027.

Northeast G&P

Adjusted EBITDA rose 7.3% to $542M, the highest of the seven quarters the release shows, on the same rich-gas mix that drove the second quarter. Consolidated gathering volumes were 4.14 Bcf/d against 4.10, but plant inlet volumes rose 9.5% to 2.08 Bcf/d and NGL production rose 13.3% to 170 Mbbls/d. Gathering, processing and fractionation revenue rose to $449M from $421M. Proportional EBITDA from equity-method investments rose to $169M from $161M on higher volumes and escalated rates at Appalachia Midstream and new minimum-volume revenue at Blue Racer. Commodity margins fell to zero from $6M after a restructured gas purchase agreement expired in 2025, and operating costs fell to $111M from $114M.

"Our Northeast G&P business grew $37 million or 7% primarily due to higher volumes across our major liquids rich and dry gas system."
— John Porter, Chief Financial Officer

Assessment: two consecutive quarters of 7% to 8% growth from the segment we called in May the weakest growth line in the portfolio. The growth is volume and rate on existing assets, with almost no capital: Northeast spent $47M of capital and equity investment in the quarter. In August management said it had been somewhat conservative on Northeast growth, and on this evidence that conservatism is a source of upside to the 2026 guide rather than a risk to it.

West

Adjusted EBITDA rose 9.3% to $401M, but Momentum contributed $40M of that in 28 days, so West excluding the acquisition fell 1.6% to roughly $361M. Haynesville service revenue rose $87M on Momentum and a full quarter of Louisiana Energy Gateway, while the Mid-Continent and Permian divestitures removed $11M. Segment operating and administrative costs rose to $215M from $150M, including $33M of Momentum acquisition costs that are excluded from adjusted EBITDA and Momentum's own operating costs. Proportional EBITDA from equity-method investments fell to $28M from $36M after the Brazos Permian II sale in June. Gathering volumes rose 16.9% to 7.18 Bcf/d, almost all of the increase from the acquired system. Modified EBITDA of $451M also carries the $83M gain on the Permian gathering sale, which closed in August for $139M.

"The West grew $34 million or about 9% including initial contributions from the Momentum acquisition, which closed in early September, as well as other higher Haynesville results including a full quarter from our Louisiana Energy Gateway pipeline and other Haynesville volume growth."
— John Porter, Chief Financial Officer

Assessment: the legacy West business is shrinking slightly, by design. Two Permian sales and the Mid-Continent sale have removed earnings, Eagle Ford minimum-volume step-downs continue, and segment capital spending stayed low at $147M. The question we raised in August about the West cost line is now impossible to answer from the outside, because Momentum's costs sit in the same line. What matters from here is whether the Haynesville platform, now the core of the segment, grows into the 2027 multiple management paid for it.

Gas & NGL Marketing Services (Sequent)

Adjusted EBITDA of $4M against $11M a year ago: commodity margins of $9M, operating costs of $19M and $17M of proportional EBITDA from equity-method investments, less a $3M linefill adjustment. Modified EBITDA of $39M includes a $32M unrealized derivative gain that management removes. Natural gas sales volumes fell 8.7% to 5.95 Bcf/d and NGL volumes rose 15.5% to 201 Mbbls/d. Year to date the segment has earned $230M, all but $3M of it in the first quarter.

"Our Sequent Marketing business was a bit lower than last year with continued seasonally lower opportunities for this business."
— John Porter, Chief Financial Officer

Assessment: the third consecutive quarter that behaves exactly like the winter option we described at initiation. Nothing in this line affects the investment case until January, when it becomes the swing factor for the top of the 2026 range and the starting point for 2027.

Other

Adjusted EBITDA fell 31.1% to $62M as the January sale of the South Mansfield upstream interests annualized. Net realized product sales fell to $128M from $151M, and operating costs rose to $90M from $71M on higher incentive compensation and Wamsutter upstream operating expense. Service revenue rose to $13M from $4M.

"Finally, our other segment, which includes our upstream businesses, was down about $28 million primarily due to our divestiture of the upstream Haynesville assets which closed in January 2026."
— John Porter, Chief Financial Officer

Assessment: the decline is planned and disclosed. Year to date the segment has earned $209M against $272M, and it is on course for roughly $270M to $280M for the year, below the ~$295M we carried in August. The unexplained item is capital: the Other segment spent $454M in the quarter against $88M in the second quarter and $65M a year ago, and neither the release nor the call explained it.

Operating Statistics

Two lines deserve isolating. MountainWest transportation volumes rose 27.3% on flat contracted capacity, the clearest sign in the release of the Rockies storage and transport demand that underpins the Clay Basin expansion announced this quarter. And the West gathering figure jumped 1.15 Bcf/d sequentially, from 6.03 to 7.18 Bcf/d, with Momentum in the average for only 28 of 92 days. That implies the acquired system gathers several Bcf/d at a full-quarter rate, which is consistent with the 6 Bcf/d of gathering capacity Williams described at announcement. The fourth quarter will show the full effect.

Key Topics & Management Commentary

Overall Management Tone: the most settled of the three calls we have covered, because for once there was little to defend and a list of delivered items to read out. Management was specific on what had landed and deliberately unspecific on what comes next: the next power projects, their financing and the cost of a second partnership were all framed as options rather than plans. The one exchange that aged badly within a day was the reassurance that the election was not a gating issue for the power backlog.

1. Socrates Completes This Week, Ahead of Schedule and Within Budget

Socrates South, the first 200 MW phase, ran its first full quarter. Socrates North, which brings the New Albany, Ohio facility to its full 556 MW, completes commissioning this week. In August that second phase was guided to the fourth quarter without a date; the release now says mid-November, and the call said later this week. The power purchase agreement on Socrates South is accounted for as an operating lease, with income recognized in service revenues on a straight-line basis over a 10-year noncancellable term and a customer option to extend by five years.

"Later this week and, again, ahead of schedule and within budget, our teams will place the second and final phase of Socrates into service. Socrates shows that when customers need comprehensive energy solutions, combined with speed to power, Williams is a partner that will deliver."
— Chad Zamarin, President and Chief Executive Officer

The operating report from the first phase was brief and positive. Management described rising utilization at the south plant and load-following performance "exactly the way we want to," and said lessons from commissioning are already being applied to the four projects under construction.

Assessment: the first project of the power franchise is completing early, with the first phase delivered in under 18 months from commercialization. That moves the proof point we flagged in August from one plant to one complete project. The remaining four, Apollo in the second half of 2027, Aquila across the second half of 2027 and the first half of 2028, and Socrates the Younger and Neo in the second half of 2028, are reported on schedule. Execution risk on the committed program is now the smallest risk in the power thesis; the larger ones sit in what is not yet committed.

2. Momentum Closed a Month Early, and the First Numbers Arrived

Williams closed the Momentum Midstream acquisition on Sept. 3, a month ahead of schedule. The quarterly filing puts total consideration at approximately $5.1B, paid in cash and 26,874,496 Williams shares, and separately records $869M of NG3 secured variable-rate debt due 2029 that Williams retained. The preliminary purchase price allocation records $5,763M of property, plant and equipment, $490M of customer-relationship intangibles amortized over 20 years, $898M of assumed long-term debt at fair value, and $450M of noncontrolling interests in the partly owned NG3 pipeline. In its first 28 days the business contributed $113M of revenue and $40M of Modified EBITDA.

"On September 3, we closed our acquisition of Momentum Midstream, 1 month ahead of schedule. And the Momentum assets integrate seamlessly into our existing Haynesville position and tie into our Transco footprint along the Gulf Coast."
— Chad Zamarin, President and Chief Executive Officer

The filing also carries the first historical figures for the target, in pro forma form. Momentum's revenue was $327M in the third quarter of 2025 and $1,063M in the first nine months of 2025. For 2026 before the close, from Jan. 1 to Sept. 2, it was $1,293M, which on a daily-rate basis is roughly 36% above the 2025 pace. Pro forma net income attributable to Williams from the business was $70M for that 2026 period against $7M for the first nine months of 2025. On synergies, management said capture is running ahead of plan, against an internal assumption that synergies would start in 2028.

"We had assumed that synergies would start to materialize in 2028, but I can tell you the team is focused on capturing you know, every ounce of that now and as early as possible."
— Chad Zamarin, President and Chief Executive Officer

Assessment: this answers part of the disclosure gap we raised in August. Revenue growing at roughly a third per year is a better trajectory than the 8.5x forward multiple required, and the September run rate of $40M in 28 days annualizes to roughly $520M against about $645M of 2027 EBITDA implied by the headline multiple. That gap is the growth management paid for, and one month of data neither proves nor disproves it. Two caveats remain. Revenue at a gathering business includes commodity pass-through and is not a margin measure, and there is still no historical EBITDA. The filing does not bridge its $5.1B consideration and the separately listed retained debt to the "up to $5.5B" headline announced in August.

3. Guidance: The Range Held, the Positioning Moved Up

Williams kept 2026 adjusted EBITDA guidance at $8.3B to $8.5B and, in the release headline, said it is tracking toward the upper half. In August the same "upper half" language applied to the base business before Momentum was layered on top. Now it applies to the total.

"For full year 2026 adjusted EBITDA, our existing businesses continue tracking toward the upper half of our $8.3 billion to $8.5 billion range."
— John Porter, Chief Financial Officer

The upper half is $8.40B to $8.50B, a midpoint of $8.45B and $50M above the $8.40B midpoint carried in August. On our arithmetic the early close explains it: Momentum's September contribution of $40M is roughly one extra month at the run rate the August guide assumed for three. Year-to-date adjusted EBITDA of $6,214M has grown 8.7%. The upper half requires $2,186M to $2,286M in the fourth quarter, growth of 7.5% to 12.4% against $2,033M. With a full quarter of Momentum at the September rate, about $130M, the base business needs roughly 3.5% growth at $8.45B. The base grew 4.1% in the third quarter excluding Momentum.

Assessment: a quiet upgrade worth about $50M, not a raise, and the right posture with Sequent's winter still ahead. The base business is tracking a little ahead of the second-half pace the August guide implied, and the fourth quarter adds Socrates North and a full quarter of Momentum. We move our 2026 estimate to $8.45B.

4. One to Two More Power Projects by Year End, Bigger and Longer

In August management promised more power commercializations before year end. This quarter it put a number on the promise and described what the next projects look like: new hyperscaler counterparties, significantly longer contract terms, and projects scoped from the start to scale in phases rather than sized only for speed to power. Pressed on whether "1 to 2" signals uneven confidence, management said it does not time announcements to calendar dates and announces only with a signed agreement, permits ready and sites secured.

"On that note, we expect to announce 1 to 2 more power projects by year's end, so stay tuned. And we expect those projects to expand our customer base with additional high quality counterparties along with significantly longer contract tenures."
— Chad Zamarin, President and Chief Executive Officer

Management also conceded that later projects may carry less of the excess capacity the first five were built with, which is part of the upside the first five offer beyond their contracted economics.

Assessment: three improvements on the first five projects are on offer: a second counterparty, a longer term than 10 to 12.5 years, and a phased scale that makes a single announcement worth more. Each one reduces a risk the bear case leans on, especially customer concentration. The trade-off is that tighter engineering may lower the upside beyond a 5x build multiple. We would accept that trade for a longer contract and a new customer. Eight weeks remain for the promise to land, and it now lands into a changed political map (Topic 10).

5. A Pay Plan That Says 10% to 16%

The quarterly filing disclosed a one-time Five-Year Performance Award for the five senior officers, approved on Oct. 29. It pays nothing unless adjusted EBITDA compounds at 10% or more from 2026 through 2030 and debt to adjusted EBITDA is at or below 4.0x on Dec. 31, 2030. Payout is target at a 10% CAGR, rising on a straight line to twice target at 16%. Target values total $29.7M, with $8.8M for the chief executive, and the maximum is double. The award is separate from the regular plans, in which management said return on invested capital remains the most important metric.

"I will tell you, everybody on our management team is super focused on winning at a 16% level. And, you know, that could include some bolt ons along the way. But I think, you know, as we focus in on our plan, we are most focused in on our backlog."
— John Porter, Chief Financial Officer

Assessment: the award does two useful things and one that needs watching. It puts management's pay behind growth well above the public 11%-plus target, and it caps leverage at the end of the period at 4.0x, which limits how much of the growth can be bought with debt. But the metric is consolidated adjusted EBITDA. Acquisitions count toward it, as the chief financial officer said, and so does 100% of the power projects of which Williams owns 51%. A per-share or attributable measure would have aligned the award more closely with shareholders. Target pay also starts at 10%, a point below the public target, so the plan pays out on a result the company would itself call a miss.

6. Funding the Next Wave: The 6.35% Is Gone

Year-end leverage is guided to about 3.75x on a full-year Momentum run rate, before any new power project, which leaves roughly $2B of headroom to the 4.0x internal ceiling to cover the initial spend on projects announced by year end. Beyond that, the funding answer is open. Management said the cost of a second partnership has risen since the July deal with the move in Treasury yields, named three options, and said 2027 financing will depend on the scale and speed of the next projects.

"Definitely, prices have moved a bit from the, you know, the 6.35% that we got on the deal with Blackstone back in July and, you know, just the underlying movement in the 10-year treasuries has caused there to be a shift in the cost of capital for that."
— John Porter, Chief Financial Officer

The options named were another partnership, insurance capital, and "perhaps running the balance sheet a little bit harder for a short period of time before all the earnings kicks in." Management committed to give financing color alongside any large, phased project announcement. Of the $5.34B committed by the partner, $3.74B arrived in July and $1B in October, leaving about $0.6B due by early 2027.

Assessment: Bear 1 stays contained for the five committed projects, which are funded. The next wave reopens the question, at a higher cost. The phrase that matters is running the balance sheet harder: with quarter-end leverage at 3.80x and the pay plan's ceiling at 4.0x for 2030, the room for that is a matter of a year or two, not a strategy. We would treat any announcement of a large phased project without a named funding source as a negative, whatever the project's build multiple.

7. $5.9B of Equipment on Order

The quarterly filing shows construction and equipment commitments of about $5.9B at Sept. 30, a significant portion of it long-lead power equipment backed by customer reimbursement if an order is cancelled. Year-end leverage and the growth capex guide both exclude certain reimbursable long-lead equipment, which stood at $857M of cash purchases in the leverage calculation, up from $777M in June.

"We have done it with back stop agreements. We have done it without really exposing the company to the risk of loss associated with that equipment."
— John Porter, Chief Financial Officer

Management said the equipment on order would support the entire guided backlog and more, with deliveries layering in through the early 2030s, and that there is no direct link between the $5.9B and the 6 GW-plus power opportunity set.

Assessment: the equipment book is the asset that makes the power backlog credible, because turbines and balance-of-plant are the binding constraint in this market. The reimbursement backstops limit the downside if a customer walks. The cost is presentational: a growing pool of capital sits outside both the leverage ratio and the capex guide until a project is sanctioned, and the leverage adjustment alone is worth about 0.1x.

8. Transco and Storage: Three More Expansions

The organic queue added three projects. Southeast Supply Enhancement II, a follow-on to the largest project in the company's history by earnings, starts at 150-plus MMcf/d under contract with a utility, against an open-season scope of up to 390 MMcf/d with a second customer in negotiation. Garden Connector was upsized by 84 MMcf/d to 144 MMcf/d one quarter after its first agreements. And a 9.4 Bcf expansion of the Clay Basin storage field on MountainWest was commercialized. Four pipeline projects totalling 413 MMcf/d were placed in service. The quarterly filing dates three of them, Gillis West, the Huntingdon Connector and the Ryckman Creek Lateral, to late October, after the quarter closed, so they do not contribute to these results.

"As we have announced right now, we have got, a 150 a day contracted on that with is with a high credit, utility counterparty. We did have a non. Binding open season on that, which articulated the scope up to 390."
— Larry Larsen, Chief Operating Officer

Management expects the second phase at a 4x to 6x build multiple and said larger Transco expansions for the 2030s are under discussion. The quarterly filing keeps Southeast Supply Enhancement's full in-service at as early as the third quarter of 2027; the early partial in-service discussed in August was not repeated.

Assessment: this is Bull 1 working as designed: brownfield additions on a corridor no one else can build, at build multiples that would be extraordinary anywhere else. Storage is becoming a second leg of the same argument, with contracted rates rising, Pine Prairie Phase IV approved in May and Clay Basin added this quarter. The late-October in-service dates also mean the fourth quarter gets a modest lift the third quarter did not.

9. NESE: A Court Remand, and Permitting Reform as the Fix

On Sept. 8 the Third Circuit vacated New Jersey's Clean Water Act water-quality certification for the Northeast Supply Enhancement project and remanded it to the state. The project is fully subscribed and guided to in-service as early as the fourth quarter of 2027. Management said the remand does not jeopardize the project, that the financial effect is immaterial, and that timing for next winter is at risk.

"And, frankly, what this does is it is not gonna have an impact on us from a financial forecasting perspective. We feel really good about the need for the project. But, you know, it could put at risk our ability to get it online in time for next winter."
— Chad Zamarin, President and Chief Executive Officer

Management linked the outcome to the permitting bill before the Senate, which it said would have protected the certificate, and pointed to a cloture vote scheduled for Nov. 9.

Assessment: NESE adds 400 Mdth/d and has never been a large part of the earnings case, so the financial claim is credible. The point is the pattern: a project with federal approval, signed customers and approved state permits lost one of those permits in court, and management's remedy is legislation. The cloture vote now happens after an election that changed control of Congress for the next session, which makes its outcome harder to call than it looked on the call.

10. The Election, Asked About on the Call and Priced the Next Day

The call took place on election day. Asked whether the timing of the projects closest to commercialization is sensitive to the governor races, with Texas named, management said the election is not a gating item. The chief executive closed the prepared remarks by urging listeners to vote, because government decisions "shape whether America can build the infrastructure it needs."

"I would not say that is a gating issue for any of the discussions that we have been having. I think everybody's keeping an eye on the election and you know, looking forward, I would say, to getting through. An election cycle."
— Chad Zamarin, President and Chief Executive Officer

The results arrived overnight. Democrats won both chambers of Congress. Texas elected a governor who promised to halt new data centers. Ohio elected Amy Acton, who campaigned on a conditional moratorium on new data centers unless they meet conditions that include union construction labor, covering all of their own utility costs, environmental standards and community benefit agreements. Her running mate said that ideally data centers would bring their own energy source. Four of the five committed power projects are in Ohio, 2,068 MW of the 2,588 MW committed. The quarterly filing states that Socrates has its Ohio Power Siting Board approvals and that the in-service dates for the other four projects assume timely receipt of permits.

Assessment: this is the new risk in the thesis, and we add it as Bear 5. Read carefully, the Ohio platform is not hostile to the Williams model: behind-the-meter generation is the textbook way for a data center to cover its own power costs, and that is close to what the campaign said it wanted. But the conditions apply to the data centers Williams serves, the incoming administration will shape the state's regulators, and the in-service dates for three Ohio projects assume timely receipt of permits. The risk falls mainly on the next wave of projects and their location, not on the contracted cash flows of the first five. Management's answer on election day was right about the gating question and wrong about where the market's attention would be the next morning.

11. Line 200 and the Woodside LNG Capacity

Line 200, the 3.1 Bcf/d pipeline to the Louisiana LNG terminal, has its permits and is in construction; the filing guides in-service as early as the second quarter of 2028, and the chief operating officer put it at the end of 2028. Williams also holds 1.5 million tonnes a year of LNG offtake from the terminal. Asked whether that capacity has gained value as the world pays more for supply security, management agreed and said it would likely term it out at a fixed margin rather than hold international price exposure, and offer it to Haynesville producers as a service.

"But I would expect, again, yes, that capacity has certainly increased in value, but our strategy will likely be to term out that capacity, make sure we capture fixed margin."
— Chad Zamarin, President and Chief Executive Officer

Assessment: the right instinct. An infrastructure company that wants an infrastructure multiple should not carry an open LNG position, and termed-out capacity paired with Momentum's gathering is a differentiated offer to producers. Management also said 75% of the drilling inventory behind the Momentum system is economic at $3 gas and virtually all at $3.75, which is the supply-side case for the acquisition in one sentence.

Guidance & Outlook

Every line of the guidance bridge is identical to August, so the per-share figures above are midpoint to midpoint and the only movement is in the language. The release headline and the chief financial officer both place the year in the upper half of the adjusted EBITDA range. Growth capex was held pending the next power announcements, with management saying the 2027 financing plan will be guided by "the scale and speed of our next power projects." It also said that before the earnings step-up of 2028 it expects "sector leading growth in 2027."

One bridge line is already stale. Noncontrolling interests and preferred dividends are carried at $220M for the year. Year to date the two total $165M, which leaves $55M for the fourth quarter against $68M in the third, and the fourth quarter adds Socrates North and a full quarter of the NG3 minority interest. On our estimate the line lands nearer $235M, which costs a little over a cent of adjusted EPS. Interest expense, by contrast, looks right: the bridge implies $403M in the fourth quarter after $385M in the third, consistent with a full quarter of the September notes and the retained Momentum debt.

Implied fourth quarter: year-to-date adjusted EBITDA of $6,214M grew 8.7%. The $8.40B midpoint requires $2,186M in the fourth quarter, up 7.5% from $2,033M; the upper-half midpoint of $8.45B requires $2,236M, up 10.0%; the $8.5B top requires $2,286M, up 12.4%. A full quarter of Momentum at its September rate is worth about $130M, so the base business needs roughly 3.5% growth to reach $8.45B, against 4.1% in the third quarter. Adjusted EPS of $1.77 year to date means the unchanged $2.35 midpoint implies $0.58 in the fourth quarter, up 5.5% from $0.55.

Street at: above the company's own per-share bridge. Published full-year 2026 adjusted EPS estimates sit at $2.37 to $2.42 against the $2.35 midpoint, and the one published fourth-quarter estimate of $0.63 is five cents above the $0.58 the bridge implies. Published 2027 estimates of $2.67 to $2.68 imply roughly 14% growth on the 2026 midpoint. No published adjusted EBITDA consensus exists for the year, so the guidance range remains the honest anchor.

Guidance style: conservative in the bridge and confident in the framing, the same posture as August. Management again declined to count on an early winter for Sequent and left the range intact with two months to go. The fourth-quarter EPS gap between the Street and the bridge is the item most likely to produce a headline miss in February, and it is a gap in the Street's model rather than in the company's guidance.

Analyst Q&A Highlights

Who the Next Power Customers Are

The first question of the call went to the 1 to 2 projects promised by year end and whether they broaden the counterparty base. The answer was unusually direct: yes, new customers, still large hyperscalers, with contract terms "significantly" longer than the 10 to 12.5 years on the first five, and project designs that show how capacity scales over time.

Q: "And 1 to 2 more, you know, deals expected by year end, Is how should we think about, I guess, customer conversations here? Is this broadening out your counterparties?"
— Jeremy Tonet, JPMorgan Securities

A: "We are only focused on, you know, really the highest quality hyperscalers and projects that we think fit really well within our strategic capabilities. And so, you know, I mentioned in the prepared remarks that, you know, we expect to expand and diversify the customer base. We will continue to be, you know, large high quality customers. And we also expect contracts to significantly extend in term length."
— Chad Zamarin, President and Chief Executive Officer

Assessment: this is the most valuable commitment of the call, because it addresses the one risk in the power franchise that execution alone cannot fix. Five projects with one primary counterparty is a customer-concentration problem. A second named hyperscaler on a longer contract converts the backlog into a portfolio. The commitment has a date attached and is checkable within weeks.

Whether the Governor Races Gate the Power Backlog

Asked on election morning how sensitive the timing of the projects nearest to commercialization is to the governor races, with Texas singled out, management said the election is not a gating issue for any discussion under way. It added that Williams sites projects with supportive communities, and that Texas will be important to data-center infrastructure.

Q: "So I guess with Election Day, here upon us, you know, when you when you look across your power backlog, particularly the projects that are closer to commercialization, I guess, how sensitive is the timing of those projects to the to the elections and specifically the governor races? I guess, And I am really curious about Texas given the size of the opportunity there and whether that has any bearing on your time lines and backlog."
— Praneeth Satish, Wells Fargo

A: "Yeah. I would not say that is a gating issue for any of the discussions that we have been having. I think everybody's keeping an eye on the election and you know, looking forward, I would say, to getting through. An election cycle."
— Chad Zamarin, President and Chief Executive Officer

Assessment: the question was asked about Texas and the answer was about gating, and within a day both turned out to matter less than Ohio, which nobody named. The answer may still prove right for projects already in negotiation. But investors are now pricing the political risk to the power backlog, and management has given them nothing specific to price it with.

The 16% End of the Pay Plan

A question on the new performance award asked what it would take to reach the 16% top of its range and how much of that would be acquired. Management said the whole team is focused on 16%, that bolt-on acquisitions could be part of it, and that the backlog is the main route. On leverage, it pointed out that the award pays nothing above 4.0x.

Q: "That low end is congruent with the 11%+ target, but I wanted to dig in a little bit more on that high end of 16% I suspect that is a deliberate and achievable target by 2030. Wanted to understand, I guess, what it would take to achieve that how much M&A is contemplated versus organic?"
— Spiro Dounis, Citi

A: "I will tell you, everybody on our management team is super focused on winning at a 16% level. And, you know, that could include some bolt ons along the way. But I think, you know, as we focus in on our plan, we are most focused in on our backlog."
— John Porter, Chief Financial Officer

Assessment: a candid answer, and the candour cuts both ways. Management did not hide that acquisitions count toward the metric that sets its pay, which keeps Bear 4 open. It also said return on invested capital remains the most important metric in the regular plans, and that it does not want to lose ground on cash returns. The 4.0x leverage condition is the real discipline in the award.

What the $5.9B of Commitments Means

A follow-up asked how $5.9B of equipment commitments disclosed in the quarterly filing relates to the 6 GW-plus power opportunity set and how many orders have converted into sanctioned projects. Management called the disclosure murky, said the total aggregates commitments that carry cancellation terms and customer backstops, and said there is no strong link to the gigawatt figure.

Q: "Looks like it pointed out almost $6 billion of long lead equipment purchases for projects not sanctioned yet. And so I am trying to reconcile what that means with respect to your 6-gigawatt-plus opportunity set."
— Spiro Dounis, Citi

A: "Those 10 Q disclosures can be sometimes a little murky. In this case, I think you are talking about the $5.9 billion that is referred to in terms of commitments. ... As far as the direct sort of relationship between the 5.9 billion and the 6 gigawatts, I would not think that there is."
— John Porter, Chief Financial Officer

Assessment: the conversion rate was not answered. Management did say the equipment on order covers the full guided backlog "and even a bit more," with deliveries into the early 2030s, which is a stronger statement about the backlog than the gigawatt figure itself. The disclosure would be more useful if it separated equipment for sanctioned projects from equipment held for the backlog.

NESE After the Third Circuit

A question on Northeast Supply Enhancement drew the general counsel's first appearance on a call we have covered. The court remanded the New Jersey certification, and in October denied rehearing. Management said nothing in the remand jeopardizes the project, that it is reevaluating its path with the state, and that the financial forecast is unaffected.

Q: "Can you just catch us up on where NESE stands right now, kind of next steps to watch? You are feeling about the timeline?"
— John Mackay, Goldman Sachs

A: "I am sure you guys are aware, the Third Circuit remanded our New Jersey certificate, but there is nothing in that remand that is really jeopardized the project. We are reevaluating right now The direction we are gonna take"
— Lane Wilson, Senior Vice President and General Counsel

Assessment: the answer separates two things that are worth keeping separate: the project survives, and its timing does not. The chief executive's follow-up conceded that next winter's in-service is at risk. The financial impact is small; the precedent of a fully permitted, fully contracted Transco expansion losing a state permit in court is not, and it is one reason we add permitting to the new bear point.

The Cost of a Second Partnership

A question asked whether the joint-venture structure is still available for the next power projects given that markets have moved since July. Management confirmed the cost has risen with Treasury yields, called a second partnership one option among several, and named the alternative of running leverage higher for a period before the 2028 earnings arrive.

Q: "I just wondered if what the ability and appetite is to use the joint venture structure for future power projects. I asked the question because maybe the market has moved a tiny bit since the first agreement."
— Robert Catellier, CIBC Capital Markets

A: "Definitely, prices have moved a bit from the, you know, the 6.35% that we got on the deal with Blackstone back in July and, you know, just the underlying movement in the 10-year treasuries has caused there to be a shift in the cost of capital for that."
— John Porter, Chief Financial Officer

Assessment: no number was given, and none was needed to make the point: the 6.35% was a price for a specific moment and a specific set of projects, as we argued in August. The more consequential part of the answer was the leverage option. It is a legitimate bridge for a company whose earnings step up in 2028, but it is the option that would reopen Bear 1.

Whether the Next Projects Carry the Same Spare Capacity

The last question of the call asked whether the next, larger projects will be over-engineered to the same degree as the first five, whose excess capacity is a source of upside beyond their contracted economics. Management said the economics will be preserved through efficiency, and agreed that later projects may carry less redundancy as it learns real load profiles.

Q: "On these longer term scaling projects, prospective opportunities. Would they be as proportionately overengineered as the legacy projects?"
— Craig Shere, Tudor, Pickering, Holt & Co.

A: "I do think we will learn over time how to be more efficient with excess capacity. So I do think that is an opportunity for us to. As we operate and learn load profiles, you know, we may be able to have less of a redundancy from an efficiency perspective. So I do think that is a fair you know, observation."
— Chad Zamarin, President and Chief Executive Officer

Assessment: an honest concession. The contracted returns on new projects should hold near the 5x build multiple, but the optionality that the bull case attaches to spare megawatts at the first five sites will be smaller at the next ones. We do not value that optionality explicitly, so the valuation is unaffected; the narrative of ever-improving power economics is slightly less tidy.

Momentum Integration After Two Months

A question asked for early integration lessons and whether synergies or problems had surfaced. Management reported no surprises, performance in line with the acquisition model, synergies arriving earlier than modelled and commercial interest from existing and new customers.

Q: "Can you share some early integration learnings Any unexpected synergies or challenges you have encountered? And just how the integration is progressing to your original plan?"
— Elvira Scotto, RBC Capital Markets

A: "Integration's off to a really strong start. Really no surprises after the first 2 months of ownership."
— Robert Wingo, Executive Vice President, Corporate Strategic Development

Assessment: two months is too early to grade an acquisition, and management said so. The useful evidence is in the filing rather than the answer: the business grew revenue at roughly a third a year before the close, and earned $40M in its first 28 days under Williams. Synergies were still not quantified, a point we flagged in August that remains open.

What They're NOT Saying

1. Where the next power projects are. The call promised 1 to 2 projects by year end and named their counterparty profile and contract length, but not their state. After an election that changed the governors of the two states most discussed in data-center siting, location is now the first question investors will ask, and the 1 to 2 announcements will answer it.
2. Which Ohio permits are still outstanding. The filing says Socrates has its Ohio Power Siting Board approvals and that the in-service dates for the other four projects assume timely receipt of permits. Three of those four are in Ohio. Which approvals remain, and from whom, was not stated.
3. A standalone power line. There is still no power caption in the release. The filing sizes the year-over-year increase in power revenue at $53M, including reimbursable fuel, and the segment's other fee revenue line moved by the same amount. Power EBITDA, the partner's share of it and Socrates' margin are not disclosed.
4. Momentum's historical EBITDA. The filing added pro forma revenue and net income, and the first month's Modified EBITDA. It still gives no historical EBITDA, no synergy figure and no bridge between the $5.1B consideration it reports and the "up to $5.5B" in the August announcement.
5. An updated noncontrolling-interest assumption. The bridge is unchanged at $220M although the third-quarter run rate implies more. It is a small amount this year. For 2027, with three more power projects entering service and the partner holding 49% of each, it is the line that translates consolidated EBITDA into per-share value, and no figure was offered.
6. Cogentrix, for the third quarter running. Expected "later this year" on the first-quarter call, the divestiture has not occurred. The investment is carried at $261M, down from $289M in June, and was not mentioned in the release or in management's answers.
7. The Other segment's capital. The segment spent $454M of capital in the quarter and $618M year to date, against $165M in the first nine months of 2025. The release does not explain it.
8. The early in-service at Southeast Supply Enhancement. In August the chief operating officer said part of the pipeline could enter service at the start of 2027. The filing keeps full in-service at as early as the third quarter of 2027, and the call did not revisit the early start.
9. Why West excluding Momentum declined. The legacy West business fell about 1.6% year over year, and West was the only segment to miss its estimate by more than a few million dollars. Divestitures and Eagle Ford step-downs explain part of it. Management did not discuss it.

Market Reaction

• Pre-print setup: WMB closed at $71.19 on Nov. 2, up 18.4% year to date against the S&P 500's 10.2%, up 0.9% over the trailing 30 days and up 23.0% over the trailing 12 months. The 52-week closing range was $56.51 to $79.40, so the stock went into the print 10.3% below its best close of the past year.
• Print day: the stock fell 4.3% on Nov. 2, from $74.35 to $71.19, before the release came out after the close. The S&P 500 fell 0.5% and a large pipeline peer fell 3.4% the same day, so the pre-release decline was sector-wide.
• Reaction session (Nov. 3): the stock opened at $69.30, down 2.7%, traded a range of $69.07 to $74.32 and closed at $72.47, up 1.8% or $1.28. The S&P 500 rose 0.4%.
• Volume: 11.7 million shares against a 30-day average of 7.4 million, or 1.6x normal.
• Next session (Nov. 4): the stock fell 6.4% to $67.86 on 20.5 million shares, 2.8x normal, while the S&P 500 rose 1.6%. Midstream peers fell 3% to 5%, the energy sector fund fell 4.7%, and the large AI-power names fell 7% to 14%.
• Since our last note: from the Aug. 4 close of $71.51, WMB was up 1.3% to the reaction close, against the S&P 500 down 2.2%. At the Nov. 4 close it was down 5.1%, against the index down 0.6%.

For the third consecutive quarter the reaction session took the same shape: a gap down at the open and a recovery through the 9:30 a.m. call. The opening gap most likely reflected sector weakness carried over from the day before and an EPS figure a cent short on two of the five panels. The recovery followed the call's sequence: Socrates finishing early, the year-end power commitments and the upper-half positioning. By the close the stock had outperformed the index by about 1.4 points, a modest but genuine response to a quarter with no new problems.

The next day erased that and more, and none of it came from the company. Overnight, Democrats won both chambers of Congress, Texas elected a governor who promised to halt new data centers, and Ohio elected a governor who campaigned on a conditional moratorium on them. AI-power equities sold off hard, and WMB traded partly with them: its 6.4% decline sits between the 3% to 5% losses in midstream and the 7% to 14% losses in power generation and equipment. That placement is the market's own answer to a question this coverage has asked since initiation: how much of Williams' value is the power franchise. On Nov. 4 the answer was roughly half the distance between a pipeline and a power producer.

We think the move prices the risk to the next wave of power projects reasonably and the risk to the committed program generously. The five committed projects have signed, primarily fixed-price agreements, one is completing this week, and the four others are on schedule. Ohio's governor-elect proposed conditions on new data centers, the reporting on her platform does not address existing contracts, and her running mate described data centers bringing their own energy source as the ideal. What the market has priced correctly is that the next commercializations, and the growth beyond the contracted 11%, now depend on state politics that turned less favourable overnight.

Street Perspective

Debate: Is Williams a Pipeline or an AI-Power Stock?

Bull view: the bull case is that the Nov. 4 sell-off confused a contracted infrastructure company with merchant power producers. The power projects are behind-the-meter plants on 10 to 12.5-year primarily fixed-price agreements, they do not depend on wholesale prices or grid interconnection queues, and the permitting bill before the Senate would require data-center developers to pay the electricity costs of their facilities, which is what a behind-the-meter plant already does.

Bear view: the bear camp argues that the stock has carried a power premium for a year, that 80% of the committed megawatts sit in a state whose incoming governor campaigned on conditions for new data centers, and that the backlog beyond the five committed projects is precisely what the market had been paying for above a pipeline multiple.

Our take: both are partly right, and the distinction is between the committed and the uncommitted. The contracted projects are among the least politically exposed power assets in the market. The next wave is not, and the growth above 11% depends on it. At 13.1x 2027E adjusted EBITDA the shares now trade about where a pipeline company with a funded 11% contracted growth rate should, which means the market is assigning little to the uncommitted backlog. That is a reasonable entry point, not a reason to reduce exposure.

Debate: Does the 10% to 16% Pay Plan Raise the Bar or Lower It?

Bull view: the constructive read is that the board has put five years of senior pay behind growth well above the public target, with double payout at 16%, and conditioned all of it on leverage ending at or below 4.0x. Management teams do not choose a 16% maximum they think unreachable.

Bear view: the skeptics note that target pay begins at 10%, a point below the public 11%-plus target, that the metric is consolidated EBITDA rather than a per-share or attributable measure, and that the chief financial officer said acquisitions could count toward the top end. A plan built that way rewards buying growth and consolidating partner-owned assets.

Our take: the leverage condition does more work than either side credits, because it limits how much of the growth can be bought with debt. The metric is the weakness. We would watch noncontrolling interests and share count alongside EBITDA, and grade the plan's results on adjusted EPS and AFFO per share, which the award does not use.

Debate: How Will the Next Wave Be Funded?

Bull view: the optimistic case is that Williams has $2B of headroom to its 4.0x ceiling, partner capital that has proved deep and competitive, insurance capital interested in the same structure, and an earnings step-up in 2028 that cures the leverage problem by itself.

Bear view: the skeptical case is that operating cash flow of $4.48B in the first nine months covered less than 80% of $5.68B of capital expenditure, that leverage rose to 3.80x at quarter end, that the partner's cost has gone up with Treasury yields, and that management has now named running the balance sheet harder as an option.

Our take: the committed program is funded and the question applies to new projects only. A larger phased project announced with a partner at a modestly higher cost than 6.35% would still be accretive at a 5x build multiple. A large project announced without a named source of capital would be the first real test of the leverage discipline the pay plan now enforces, and it is the scenario we would treat as a negative.

Debate: Was Momentum Bought Well?

Bull view: the bull argument is that the deal closed a month early, synergies are arriving ahead of a 2028 plan, the target's revenue was growing at roughly a third a year before the close, and three-quarters of its drilling inventory is economic at $3 gas, which secures volume growth into the Gulf Coast demand build.

Bear view: the bear argument is that revenue includes commodity pass-through, that no historical EBITDA has been disclosed, that the first month annualizes to about $520M against about $645M of 2027 EBITDA implied by the headline multiple, and that the acquisition is one reason leverage rose.

Our take: the evidence so far is mildly positive and very incomplete. The gap between the September run rate and the 2027 figure is the growth management paid for, and one month cannot test it. We continue to underwrite the acquisition at roughly 9x net of the minority interest, with no value for synergies, and will grade it on the fourth-quarter West segment, the first with a full quarter of ownership.

Model Update Needed

Valuation. At the Nov. 4 close of $67.86 on 1,252M shares, equity value is approximately $85.0B. On the reported Sept. 30 balance sheet (net debt of $32.2B, noncontrolling interests of $5.8B and preferred stock) enterprise value is about $122.9B, or 14.5x the $8.45B upper-half midpoint, against 14.4x on the corresponding basis in August. The shares trade at 28.9x the $2.35 guided adjusted EPS midpoint against 30.4x last quarter, and at 13.2x the $5.15 AFFO-per-share midpoint against 13.9x, for a 3.1% dividend yield on the $2.10 annualized rate.

The forward framework uses 2027 estimates and a pro forma capital structure, and two of its inputs move this quarter. Noncontrolling interests rise to $7.4B from $6.0B: the July contribution put $3.14B of the first $3.74B into noncontrolling interests, against our assumption of a larger share landing in capital in excess of par, and Momentum added $450M for the NG3 minority. That costs about $1.10 per share. The share count falls to 1,252M from 1,270M because Momentum required 26.9M shares rather than the ~28M we assumed and the base count was lower than we carried, which adds about $1.15. The two changes almost exactly offset. Every $1B of noncontrolling interest is worth about $0.80 per share.

Twelve-month target: $82, maintained. The base case is unchanged at 15.0x 2027E adjusted EBITDA, and the corrected inputs land it at $82.28. From the Nov. 4 close that is 20.8% price appreciation plus a 3.1% yield, a total return of roughly 24%; from the reaction-session close of $72.47 it was roughly 16%. The bear case, at 13.0x, assumes Momentum delivers only its entry multiple and the market capitalizes the power franchise as merchant generation. It now sits 1.2% below the price, against 6.1% below in August and 19.8% below at initiation. We do not lower the multiple for the election result, because the 2027 EBITDA in the framework comes from contracted, in-construction assets. If the next wave of power projects slips materially or relocates on political grounds, the risk is to the 2028 to 2030 path and to the bull case, not to the base.

Thesis Scorecard Post-Earnings

The pillars below are the standing thesis carried since initiation. None of the existing pillars changes status this quarter. One new risk is added.

Overall: the thesis is unchanged on the company and weaker on the environment. Williams delivered the August commitments that fell due this quarter, several of them early, and the base business is running a little ahead of the guide. Two August items remain open: the year-end power announcement, which is not yet due, and the early partial start at Southeast Supply Enhancement, which management did not repeat. The new risk is external: the political conditions for siting the next wave of behind-the-meter power deteriorated in the two states that matter most, on the night of the call. It does not touch the contracted cash flows in our 2027 estimate, and it does reduce the probability of the growth above 11% that the bull case and the pay plan's upper range depend on.

Action: buy on the election weakness. At $67.86 the shares trade at our bear-case value, the committed program is funded and on schedule, and the market is now assigning little to the uncommitted power backlog. In August we wrote: "The two things that would reverse this view are a Momentum integration that requires the multiple to compress on synergies that never get quantified, and a third consecutive quarter of growth capex being raised without a corresponding step-up in the contracted book." The growth-capex trigger did not fire, because growth capex guidance was held this quarter. The synergy trigger has not fired but stays live, because synergies are still unquantified. We add two new triggers: a year-end power announcement that is delayed or relocated for political reasons, or a large project announced without a funding source and with leverage guided above 4.0x. Neither new trigger has fired, and the first will be tested within eight weeks.