Best Operating Quarter of the Decade, But the Earnings Step-Up Is Rented, Not Owned

Results vs. Consensus

A word on the consensus figures before the scorecard, because the dispersion is unusually wide and it changes how the beat should be read. Published Q1 2026 estimates for MPC ranged from roughly $0.72 to $1.09 of adjusted EPS. The Middle East supply disruption that lifted global cracks landed in the final weeks of the quarter, and most sell-side models were struck before it. The most widely cited mark, and the one used below, sits at $0.72. The honest characterization is not that MPC beat by 129%; it is that MPC beat a stale number by a wide margin, and beat the highest published estimate by 51%.

Year-over-Year Comparison

Sequential Comparison

The sequential picture is the one that matters for calibrating what a normal MPC quarter now looks like, and it is a step down, not a step up. Q4 2025 was the stronger print on almost every operating line. Q1 is a seasonally weak quarter that was made weaker by a deliberate turnaround pull-forward, and the margin environment still carried it to $2.8B of adjusted EBITDA.

Segment Performance

The single most important structural fact in that table is that Midstream is still the larger of the two big segments, at 53.0% of the segment subtotal. MPC is a refiner with a very large midstream business attached, and in a quarter where refining tripled its segment EBITDA, midstream still contributed more. That is the shape of the enterprise, and it is why the buyback is underwritten by something other than crack spreads.

Refining & Marketing

R&M produced $1,377M of segment adjusted EBITDA against $489M, on essentially identical throughput. Every dollar of the improvement is margin. The refining margin rose to $17.74 per barrel from $13.38, and the two controllable cost lines moved against the company by a combined $0.88 per barrel, so the net segment margin per barrel went to $5.37 from $1.91.

Utilization was 89%, identical to the year-ago quarter, but that comparison understates the achievement: MPC guided Q1 to 2,540 mbpd of crude runs (roughly 85% utilization) with $465M of turnaround expense, and actually ran 2,664 mbpd (89.2%) while spending $530M. It ran hotter than plan while doing more maintenance than plan. Management characterized the reliability performance in terms that are checkable against the operating statistics rather than atmospheric.

"This was our strongest first quarter on process safety as well as our lowest level of unplanned downtime this decade, all while completing approximately 40% of our full year planned maintenance activity." — Maryann Mannen, Chairman, President and CEO

Assessment: This is the segment doing the work and it is doing it well, but the margin input is exogenous and the cost inputs are not yet improving. The right way to underwrite R&M from here is on the $5.37 per barrel of segment EBITDA and the direction of the two cost lines, not on the $17.74 margin, because the margin belongs to the market and the costs belong to management.

Regional refining margin ($/bbl of net refinery throughput)

Regional segment adjusted EBITDA ($/bbl of net refinery throughput)

The two regional tables were newly disclosed at this level of granularity, and they are the most useful thing MPC gave investors this quarter. They also make the segment's internal dispersion impossible to ignore: West Coast earned $11.61 per barrel, Gulf Coast $6.20, and Mid-Continent $1.50. The Mid-Continent is the only region that went backwards year over year, and it went backwards on rising margin, from $13.03 to $14.10 per barrel, because refining operating cost per barrel rose to $6.20 from $4.94 and turnaround cost per barrel rose to $1.56 from $0.63. On the call the region was described as performing in line with seasonal expectations in January and February and widening out in March.

Assessment: Introducing regional EBITDA per barrel is a genuine transparency upgrade and a slightly risky one for management, because it creates a public scoreboard on which the Mid-Continent currently reads badly. That is the correct trade to make, and we would rather own a management team that publishes the number it is worst at.

Midstream

Midstream adjusted EBITDA fell 7.1% to $1,598M. Three items explain the decline: $77M of derivative losses on economic hedges, the absence of a $37M non-recurring customer-agreement benefit booked in the prior-year quarter, and the divestiture of non-core gathering and processing assets. Underlying volumes were softer across the board: pipeline throughput 5,788 mbpd versus 6,022, gathering 6,488 MMcf/d versus 6,516, natural gas processed 9,406 MMcf/d versus 9,781, and fractionated volumes 634 mbpd versus 660.

Against that, the growth pipeline is dense and near-term. Secretariat I is in service and ramping over nine to twelve months toward 1.4 Bcf/d of regional processing capacity, Harmon Creek III is on track for third-quarter startup taking Northeast capacity to 8.1 Bcf/d, and the Titan sour-gas treating expansion is expected to exit 2026 above 400 MMcf/d. MPLX is deploying over $2.4B of organic growth capital this year with roughly 90% directed at natural gas and NGL value chains, and management put a specific distribution commitment against it.

"MPLX expects to deliver 12.5% distribution growth for the next 2 years, underpinned by mid-single-digit adjusted EBITDA growth." — Maryann Mannen, Chairman, President and CEO

Assessment: This is the load-bearing wall of the MPC equity story and it wobbled slightly this quarter without cracking. A 12.5% distribution growth commitment against mid-single-digit EBITDA growth is arithmetically a coverage-ratio decision as much as a growth decision, and it is the item we will grade hardest as the 2026 projects convert from construction to cash in the second half. The market noticed the softness: MPLX units fell 2.6% on the session while MPC rose 3.2%.

Renewable Diesel

The segment swung to $38M of adjusted EBITDA from $(42)M, on a Renewable Diesel margin of $133M versus $26M. The swing had two drivers of very different quality. The durable one is a better margin environment. The non-durable one is the recognition of clean fuel production tax credits following February 2026 proposed guidance clarifying the 45Z qualification criteria, worth $32M, which management then excluded from adjusted earnings. The Martinez joint-venture facility completed a turnaround in the quarter, so utilization was depressed; management guided to low-90s utilization in Q2.

Assessment: At 1.3% of segment subtotal EBITDA this is a rounding error on the enterprise, and its main function in the investment case is optionality on a regulatory regime that has cost MPC money for two years running. FY2025 segment adjusted EBITDA was $(110)M and FY2024 was $(150)M. A quarter at positive $38M is progress worth noting and nothing worth paying for yet.

Corporate

Corporate expense of $274M against $210M is a $64M year-over-year headwind, attributed to fair-value remeasurement of outstanding performance-based stock compensation "driven by recent stock performance" and to environmental remediation expense relating to historical operations at the Martinez refinery. Q2 is guided to $240M.

Assessment: Half of this is a good problem (the compensation mark rises because the equity rose 76% over twelve months) and half of it is a real cash liability with an indeterminate tail (Martinez remediation). Neither is large enough to change a thesis. Both are large enough to note that at $274M per quarter, Corporate now consumes roughly 9% of segment EBITDA in a strong quarter and would consume considerably more in a weak one.

Key KPIs

Key Topics & Management Commentary

Overall Management Tone: Confident and, unusually for this company, specific. Prepared remarks were built around operating statistics that are checkable against the release rather than around adjectives, and the commercial organization was given far more airtime than is typical, with concrete named trades rather than framework language. Where management was least convincing was on capital-return pacing: the only pointed question of the call went unanswered in substance, and the contrast with the granularity offered everywhere else was conspicuous.

1. The supply shock: what came offline and for how long

The entire year-over-year earnings delta is downstream of a supply event that management sized explicitly, which is more than most refiners will do. The framing matters because it separates a demand story (durable) from a supply-outage story (not).

"Late in the first quarter, geopolitical events tightened global markets, disrupted trade flows, and drove global cracks higher. While estimates vary, we believe approximately 6 million barrels per day, representing close to 6% of global refined products capacity has come offline during the conflict in the Middle East." — Maryann Mannen, Chairman, President and CEO

Management added roughly another million barrels per day of Russian capacity to that figure, and was candid that the restart clock is unknowable: the timeline "remains dependent on the extent of any damage to facilities and resumption of crude flows to those refineries." Separately, management argued the situation gives US refiners a structural position because their crude supply is not the disrupted commodity.

"We are largely insulated from global crude supply disruptions given our crude sourcing comes mainly from the United States and Canada." — Maryann Mannen, Chairman, President and CEO

Assessment: Both statements are correct and the second is the more important one, because it explains why MPC's crack capture improved rather than compressed while input costs spiked globally. But 6% of global refining capacity returning is a single headline away, and management offered no scenario for what the business earns when it does. That asymmetry is the entire basis of our rating.

2. Buying second-quarter optionality with first-quarter earnings

The most consequential capital-allocation decision in the quarter was not the buyback authorization. It was the decision to accelerate maintenance into a seasonally weak quarter so that the asset base would be available in a strong one. MPC spent $530M of turnaround expense against a $465M guide and completed roughly 40% of the $1.35B full-year plan in the first quarter, while explicitly denying that scope had been reduced.

"We didn't change scope. This is not a reduction of scope at all. This was just our intent to bring that cost forward so that we are prepared as we are running strong in the second quarter, given the demand that we're seeing." — Maryann Mannen, Chairman, President and CEO

The full-year turnaround outlook was reaffirmed at $1.35B, which means $820M remains after Q1 and, against the $300M Q2 guide, $520M is implied for the second half.

Assessment: This is the highest-quality decision on the call and the one that most deserves to be underwritten. It cost roughly $65M of Q1 expense against plan and buys roughly 4.4 points of Q2 utilization into what management believes is the strongest margin environment in years. It is also a decision that only pays if the crack environment persists into Q2, which is a bet management has now made with real money rather than with commentary.

3. Capture at 99%, and the two things that kept it under 100%

Capture, the ratio of realized margin to the benchmark indicator margin, was 99% in the quarter. Management twice made the stronger claim that it would have exceeded 100% but for two identified drags.

"But if we look at just the first quarter alone, our capture would have been in excess of 100% this quarter, if not for two things that you mentioned. One, secondaries; and two, the impact of derivatives." — Maryann Mannen, Chairman, President and CEO

The derivative drag was eventually quantified, but only when an analyst asked for it directly at the end of the Q&A, and only in gross terms. The secondary-product drag was never quantified at all.

Assessment: A 99% capture in a violently volatile quarter is a strong result and management is entitled to the ex-items framing. What it is not entitled to is the free pass: derivatives are described as "a normal and ordinary course through our risk lens", which is precisely the argument for leaving them in the capture number rather than treating them as an adjustment. The reported figure is 99%, and 99% is the one we will track.

4. Crude sourcing is where the edge actually lives

The single most informative passage on the call was the commercial organization's account of what it did with the crude slate, and it is the part of the business that would survive a crack normalization. Sour crude rose to 48% of throughput from 46%.

"We more than doubled our U.S. Gulf Coast Canadian volumes in response to the rising premiums on the U.S. Gulf Coast that we've seen. And April actually is going to be a record volume of Canadian volumes for us system-wide." — Rick Hessling, Chief Commercial Officer

The list continued: increased Bakken volumes into the Mid-Continent and Bakken moved to the Pacific Northwest to displace higher-cost waterborne barrels, record volumes of local Utica and Marcellus condensate at Canton and Catlettsburg, five Venezuelan cargoes with the discipline to stop there, and roughly 10 million barrels of SPR crude bought directly from the Department of Energy for second-quarter runs. The restraint on the Venezuelan cargoes is the tell that this is optimization rather than opportunism.

"And what I would leave you with there is we would have bought more but we had better alternatives leaning into heavy Canadian." — Rick Hessling, Chief Commercial Officer

Assessment: This is the most durable competitive advantage MPC disclosed this quarter, and it is one that does not require a war to work. Inland pipeline connectivity to disadvantaged crude is an asset that pays more when waterborne barrels are bid up, and pays something in every environment. If we upgrade this name, this is the line item that will do it.

5. What the advantaged barrel costs on the way out

The yield table tells the other half of the crude-sourcing story, and management did not narrate it. Against prepared remarks describing a system running maximum diesel and jet, the disclosed product yields moved as follows.

Distillate yield was flat, gasoline yield fell 71 mbpd, and the largest single increase anywhere in the table was heavy fuel oil, up 51 mbpd system-wide. At the Gulf Coast specifically, heavy fuel oil went to 152 mbpd from 47 while gasoline fell to 539 from 598. That is the same region that more than doubled its Canadian heavy volumes.

Assessment: The most straightforward reading is that running more heavy, sour, discounted crude pushed more barrels toward the bottom of the barrel, and the crude discount more than paid for the yield give-up: Gulf Coast margin went to $17.58 per barrel from $11.75 and Gulf Coast segment EBITDA to $674M from $78M. The trade worked emphatically. But it is a trade whose economics depend on the width of light-heavy differentials, not just on cracks, and it means "max diesel" is a description of intent rather than of the realized slate. Management should be asked to reconcile the two.

6. Jet fuel as the yield project of record

The capital program's near-term theme is jet optionality, and one project landed inside the quarter. Garyville brought more than 30,000 barrels per day of incremental jet production capacity online in March, with Robinson expected to add roughly 10,000 barrels per day in the third quarter and El Paso's FCC and alkylation upgrade targeted for the second quarter. The commercial read on the timing was unambiguous.

"And boy, the Garyville project came online at the absolute perfect time for us. And it's really a classic example of our strategic capital deployment to enhance our yields and our competitiveness to create capture tailwinds." — Rick Hessling, Chief Commercial Officer

Management also noted that the Robinson design retains flexibility to revert to other distillates if jet economics deteriorate, which is the correct way to build a yield project into a volatile product spread.

Assessment: Approximately 25% of the 2026 refining value-enhancing capital budget is pointed at Garyville, against a stated hurdle of roughly 25% returns on refining projects. A 30,000 barrel per day jet uplift arriving in the exact quarter that jet-to-diesel spreads blew out is partly luck and entirely the point of holding optionality. This is the cleanest example in the release of capital converting to margin on a visible timeline.

7. Export creativity, and the first-time trades

Two of the quarter's disclosed trades had never been done before by this system, which is a useful proxy for how far the commercial organization is now willing to reach.

"We moved ULSD from LAR to Australia. First time ever we've done that. I give the team huge kudos on creativity and capturing a market opportunity, and we moved naphtha to Asia, first time we've done that." — Rick Hessling, Chief Commercial Officer

Jones Act waivers were also used to move jet from the Gulf Coast to Alaska and alkylate from the Gulf Coast to Los Angeles, and export demand was characterized as strongest in Europe and Latin America.

Assessment: First-time trades are worth noting precisely because they are not repeatable as first-time trades. What is repeatable is the capability, and MPC now demonstrably has arbitrage reach it did not have two years ago. The correct place for this in a model is a modestly higher steady-state capture assumption, not a discrete revenue line.

8. MPLX as the funding engine

The structural argument for MPC, made repeatedly, is that MPLX's distribution stream covers MPC's standalone capital program and its dividend, which leaves refining cash flow available for repurchase. Management stated this directly.

"I mentioned we look at the strength of the MPLX growth and the distribution that it provides to MPC. This covers the $1.5 billion that we've committed to capital on the MPC side and our growing dividend." — Maryann Mannen, Chairman, President and CEO

The MPC standalone 2026 capital outlook is $1.5B, split roughly 65% value-enhancing and 35% sustaining. MPLX is separately deploying over $2.4B of organic growth capital at a mid-teens return hurdle.

Assessment: This is the argument that justifies a lower cost of equity for MPC than for a pure-play refiner, and this quarter it held even as Midstream EBITDA fell 7%. The vulnerability is that the 12.5% distribution growth commitment for two years is running ahead of mid-single-digit EBITDA growth, which draws on coverage. The commitment is credible for 2026 on projects already funded. For 2027 it needs the second-half 2026 startups to land.

9. The LPG export chain gets its first anchor customer

The longest-dated item disclosed was the commercialization of the Gulf Coast fractionation and export complex, which does not enter service until 2028 and 2029.

"Building on that momentum, through an agreement with our South Korean customer, E1, we have secured long-term delivered demand for up to 40% of the volumes MPC will purchase from MPLX's new Gulf Coast fractionation facilities, which are adjacent to MPC's Galveston Bay refinery." — Maryann Mannen, Chairman, President and CEO

Management described the remaining volume as targeted at a mix of delivered contracts, free-on-board contracts, and a deliberate spot retention, and framed the two 150 mbpd fractionators plus the 400 mbpd export terminal joint venture as a growth platform rather than a fixed-size asset.

Assessment: A 40% take-or-pay-style anchor three years before startup materially de-risks a project that will consume capital through 2029, and the delivered structure captures freight margin that a free-on-board contract would surrender. It is also entirely outside the window in which this stock will be re-rated. Note it, do not pay for it.

10. Regional disclosure with compensation attached

The new regional adjusted EBITDA per barrel disclosure is not a presentational change; management tied it explicitly to how executives are paid.

"Our executive compensation, the annual cash bonus element of that. It's a portion of that, about 20% actually is focused on the delivery of being the most competitive in every region where we operate." — Maryann Mannen, Chairman, President and CEO

Assessment: Publishing the metric your bonus is measured against, at the granularity where you are weakest, is a real governance signal. It also creates an obligation: having framed regional per-barrel competitiveness as the objective, management now owns the Mid-Continent's $1.50 per barrel in a way it did not before this quarter.

11. Capital return, and the question that did not get answered

MPC returned over $1.0B in the quarter, including $750M of repurchases, at a 62% payout ratio against cash flow from operations excluding working capital. The board added a $5B authorization on top of $3.6B remaining, taking pro-forma capacity to $8.6B, or roughly 11.3% of market capitalization at the session close.

"And today, we announced an additional $5 billion share repurchase authorization, reinforcing our commitment to delivering industry-leading returns through cycle." — Maryann Mannen, Chairman, President and CEO

Pressed on whether the repurchase would be paced against the commodity cycle or executed ratably, management restated philosophy rather than method. That exchange is reproduced in the Q&A section below.

Assessment: Authorization capacity is not a plan. MPC is buying its own stock after a 55% year-to-date advance, in a quarter whose earnings were lifted by a supply outage, and the company declined to describe any framework linking repurchase pace to either valuation or cycle position. Management is entitled to keep that discretion. Investors are entitled to discount it, and we do.

Guidance & Outlook

MPC does not guide to earnings. It guides to volumes and controllable costs, which for a refiner is the more useful disclosure anyway, because it separates what management controls from what the market hands it. The Q2 guide is materially easier than the quarter just reported on every line.

Implied ramp: The Q2 guide asks for 2,795 mbpd of crude runs against 2,986 mbpd of crude refining capacity, which is roughly 94% utilization, versus 89.2% delivered in Q1. That comes with $230M less turnaround expense and $0.58 per barrel lower refining operating cost. Holding the Q1 refining margin of $17.74 per barrel flat and applying the guided cost structure to the guided volume would produce a materially higher R&M segment result than the $1,377M just delivered, before any assumption about where cracks go. This is the single most important thing in the release: the controllables alone create a large sequential step-up.

Full-year turnaround: Reaffirmed at $1.35B, unchanged. With $530M spent and $300M guided, $520M is implied across the second half. Management was explicit that the Q1 acceleration was a timing decision, not a scope reduction, which means the second-half figure should be taken as reliable rather than as a pool that can be quietly harvested.

Guidance style: Conservative on volume, honest on cost. MPC guided Q1 crude runs at 2,540 mbpd and delivered 2,664 mbpd, a 4.9% beat on its own volume guide, while overspending the turnaround guide by $65M and the operating-cost guide by $0.38 per barrel. That is the pattern of a company that sandbags throughput and does not sandbag cost. Read the Q2 volume guide as a floor and the Q2 cost guide as a genuine target.

What is not guided: There is no margin indicator, no capture guide, no full-year EBITDA or EPS framework, and no scenario analysis around the return of disrupted capacity. For a business whose entire year-over-year earnings delta came from the margin environment, the guidance package addresses everything except the variable that mattered.

Analyst Q&A Highlights

The second-quarter utilization guide and whether reliability is repeatable

The call opened on the Q2 outlook, and specifically on whether a 94% utilization guide off the back of an accelerated turnaround programme reflected genuine reliability improvement or favourable scheduling. Management separated the two: the volume step-up is scheduling, and the reliability improvement is the separate claim, evidenced by the lowest unplanned downtime of the decade. The response also introduced the two capture drags that would recur through the rest of the session.

Q: "I just want to start off on Slide 12, the second quarter outlook. And Maryann, the utilization guide looked very good here at 94%, but maybe you can unpack the numbers a little bit more, give a sense of your plans for each of the regions."
— Neil Mehta, Goldman Sachs

A: "We didn't change scope. This is not a reduction of scope at all. This was just our intent to bring that cost forward so that we are prepared as we are running strong in the second quarter, given the demand that we're seeing."
— Maryann Mannen, Chairman, President and CEO

Assessment: A direct answer to a direct question, and one that can be verified against the release rather than taken on trust. The $1.35B full-year turnaround reaffirmation is the corroborating disclosure; had scope been cut, the annual figure would have moved. This exchange is the reason we treat the Q2 volume guide as credible.

Cash-return cadence when the earnings are cycle-inflated

The sharpest exchange of the call concerned whether the current earnings environment should be treated as a windfall and, if so, how that should inform repurchase pacing. The questioner made the risk concrete by pointing to a single-day $17 move in oil on a headline about the Strait reopening, and asked whether the buyback would be paced against that risk or executed ratably. The answer restated capital-allocation philosophy without addressing method, and a follow-up seeking a one-word clarification produced no more.

Q: "Maybe just to clarify, would you pace the share buyback? Or is it ratable?"
— Doug Leggate, Wolfe Research

A: "We try to do our very best across the system to make the best decisions that we can and be as disciplined as we can when we are returning capital, Doug."
— Maryann Mannen, Chairman, President and CEO

Assessment: This is the least satisfying moment of an otherwise specific call, and it is not a small one. A company announcing a $5B incremental authorization on the same morning it reports a windfall quarter, at a record share price, should be able to articulate whether it buys more when the stock is cheap. The absence of a framework is itself information, and it is the primary reason this initiation is a Hold rather than an Outperform.

Whether capture compresses when cracks spike

A recurring structural objection to owning refiners into a crack spike is that realized capture falls as the benchmark widens, so the earnings uplift is a fraction of what the indicator implies. The question put that bear case directly and observed that Q1 would not have beaten without strong March capture. Management's response was that volatility is the input to capture rather than the enemy of it, provided the commercial organization is positioned to act.

Q: "But the other thing I also wanted to talk to you a bit about is there's a bear thesis which is floated around that as the refining crack tends to spike. There's a proportionate drop in capture. Now when I look at your 1Q results, your beat would not have been possible if March capture was not strong."
— Manav Gupta, UBS

A: "When we look at this unprecedented situation and scenario we're in, you're seeing extreme volatility. And I really want to make that point because that leads to incredible capture generation if and when we execute, and I think you're going to -- you've seen here in 1Q and you'll see going forward, our team is extremely energized in this environment and our ability to expand the crack was quite evident."
— Rick Hessling, Chief Commercial Officer

Assessment: The claim is partially validated by the print. Capture of 99% during a quarter that included a violent March is a good outcome and not the compression the bear case predicts. But the answer is qualified by its own conditional: capture generation happens "if and when we execute". That is a statement about capability, not about the structural relationship between crack width and capture, and the question as posed was not really answered.

Sizing the derivative drag

Derivatives were referenced three times as a capture headwind before anyone asked what they cost. When the question finally came, appended to an unrelated follow-up, the figures were provided without hesitation and were larger than the running commentary had implied.

Q: "And if I could just sneak one last one in for any voting of doubt, can you provide the derivative impact that you've alluded to a few times that you've incurred for 1Q?"
— Jason Gabelman, TD Cowen

A: "Jason, so from a derivative perspective, in the first quarter, we had about $500 million of unrealized losses. And if you think about midstream was about $63 million. and then the impact on margin calls to working capital was about $340 million overall use of cash."
— Maria Khoury, Executive Vice President and Chief Financial Officer

Assessment: The disclosure is complete and the answer is the correct one to give, but the fact that a roughly $500M unrealized item required an analyst to extract it, as an appended third question, is a disclosure-practice observation worth logging. Management characterized the programme as unchanged from prior quarters, with only commodity volatility different. That is credible and checkable, and the reversal claim is now a specific, gradeable commitment for the second quarter.

The Mid-Continent and West Coast diverging in opposite directions

The regional dispersion in the new per-barrel disclosure drew a question about whether inland weakness and West Coast strength were seasonal or structural, and whether both would mean-revert into the second quarter. The answer inverted the premise on the Mid-Continent: the region behaved to seasonal expectations early in the quarter and has since become the strongest market in the system on inventory tightness.

Q: "I wanted to start on what you're seeing kind of in the inland markets and West Coast and kind of we've seen a bit of inverse performance there in land was, I'd say, weaker than normal seasonality in 1Q. What do you think drove that?"
— Jason Gabelman, TD Cowen

A: "And as we look at the Mid-Con today, it is absolutely the best market we have right now within our system. We're seeing extreme tightness on gas and diesel inventories."
— Rick Hessling, Chief Commercial Officer

Assessment: This is the most actionable forward-looking colour on the call, because it is a statement about the current quarter rather than the reported one, and it is specific enough to be wrong. The Mid-Continent earned $1.50 per barrel of segment EBITDA in Q1 and carried $1.56 per barrel of turnaround cost; if the region is genuinely the system's best market in Q2 with maintenance behind it, the sequential recovery there is larger than the headline utilization step-up implies.

Product demand elasticity at higher pump prices

With product prices rising on the supply disruption, a natural question is whether demand destruction is visible anywhere in the system. The response was uniformly negative on elasticity, with an unprompted observation that the West Coast wholesale position has improved as competitors exit the market rather than deteriorated on price.

Q: "Given the upside volatility in slot prices that we've seen, I would love to get an update on your demand observations and outlook across the products within your footprint. Are there any regions or areas that you're seeing any resistance as far as demand goes, any areas where we're testing that point of inflection for demand elasticity?"
— Theresa Chen, Barclays

A: "Theresa, it's Rick. So I would say resilient is the word I would leave you with. We continue to see resilient demand for both gas and diesel across Mid-Con, Gulf Coast and the West Coast."
— Rick Hessling, Chief Commercial Officer

Assessment: A clean answer, and the West Coast wholesale share gain is the more interesting half of it. But the question asked where the inflection point is, and the answer described where it is not. Refined product demand elasticity is a lagging observation; a quarter in which no elasticity is visible is weak evidence about the quarter in which it becomes visible.

Competition for capital between refining yield projects and the midstream gas build

With MPLX directing roughly 90% of a $2.4B growth budget at natural gas and NGL infrastructure while the refining segment runs its own yield-improvement programme, the question was whether the two compete and whether refining yield gains create logistics constraints that midstream then has to solve. Management's answer was that the two capital pools are governed separately with different hurdles and different mandates.

Q: "So I guess, to sum it all up, competition for capital across logistics with the backdrop of all these yield benefits coming to the refinery level?"
— Sam Margolin, Wells Fargo

A: "As we move to the refining side, our target there, as I mentioned in the past, we're looking at returns in about 25%. We want yield optimization we want cost reduction. We want to continue to focus on those assets that are delivering the best profitability over the long term."
— Maryann Mannen, Chairman, President and CEO

Assessment: A 25% hurdle on refining capital against a mid-teens hurdle on midstream growth capital is a coherent and defensible split, and it explains why MPC's own capital budget is only $1.5B against MPLX's $2.4B. The unexamined premise in both question and answer is that these hurdles are being met; no realized-return disclosure exists against either. That is a reasonable thing to ask for in a future quarter.

What They're NOT Saying

1. Any scenario for the return of the 6 million barrels per day: Management sized the outage precisely and then declined to frame what the business earns when it reverses, beyond noting the timeline is uncertain. For a quarter whose entire year-over-year delta is that outage, an unquantified reversal case is the largest omission in the release.
2. The size of the secondary-product drag on capture: Capture was 99% and management twice said it would have exceeded 100% but for secondaries and derivatives. The derivative half was eventually quantified at roughly $500M of unrealized losses. The secondary half was never sized, in either dollars or capture points, despite being cited as one of two named causes.
3. Any repurchase pacing framework: Asked directly whether the buyback is paced or ratable, management gave no method. On an $8.6B pro-forma authorization equal to roughly 11.3% of market capitalization, at a record share price, after a 55% year-to-date advance, this is a material non-answer rather than a routine one.
4. Why heavy fuel oil yield rose 51 mbpd while the system was in maximum diesel and jet mode: The prepared remarks and the commercial commentary both described running maximum diesel and jet. The disclosed yield table shows distillate essentially flat, gasoline down 71 mbpd, and heavy fuel oil up 51 mbpd, concentrated at the Gulf Coast. Nobody asked, and management did not volunteer, how the advantaged heavy crude slate interacts with the product-yield objective.
5. The size of the divested gathering and processing assets: Asset divestiture was named as one of three drivers of the 7.1% Midstream EBITDA decline, alongside two items that were quantified ($77M of derivative losses, a $37M prior-year benefit). The divestiture contribution was left unsized, which makes the underlying organic Midstream trend unrecoverable from the disclosure.
6. Realized returns on the refining and midstream capital programmes: Management restated the 25% refining hurdle and the mid-teens midstream hurdle several times. No project-level or programme-level realized return was disclosed for anything already in service, including the completed Los Angeles utility-systems investment that was credited with improving reliability this quarter.
7. What the Martinez environmental remediation obligation totals: Corporate expense rose $64M year over year on two items, one of which is remediation relating to historical operations at Martinez. No accrual, no range, no expected duration.

Market Reaction

• Pre-print setup: MPC closed at $252.54 on May 4, which was itself the highest close of the trailing 52 weeks (52-week closing range $137.72 to $252.54). The stock entered the print up 55.3% year to date against +5.2% for the S&P 500, up 76.2% over trailing twelve months, and up 4.5% over the trailing 30 days. This was a strong tape into a strong print, which is the setup in which good numbers most often fail to move a stock.
• Reaction session: MPC opened at $256.00, a 1.4% gap, traded a $251.89 to $261.61 range, and closed at $260.51, up 3.2% or $7.97 on the day. Volume was 2.5M shares against a 2.4M 30-day average, a 1.0x day. The S&P 500 rose 0.8%.
• Positioning: The close at $260.51 is a new 52-week closing high. The intraday low of $251.89 sat marginally below the prior close, so the print was never seriously challenged during the session.
• Peer reaction: Large-cap refining comparables moved far less. Valero closed +0.7% and Phillips 66 +1.0%, against MPC's +3.2%. Smaller refiners were mixed to stronger, with HF Sinclair +3.8% and PBF +1.5%. Exxon Mobil rose 0.8%.
• Affiliate reaction: MPLX units fell 2.6% on the same session, on the softer Midstream segment result.

The shape of the move is the informative part. MPC outperformed Valero and Phillips 66 by roughly 220 to 250 basis points on a day when the crack environment was common knowledge and priced into every refiner. A rising tide would have moved the group together; it did not. The differentiated portion of the move is therefore about MPC-specific content, and there are only two candidates: the operating beat against its own volume and utilization guide, and the $5B incremental repurchase authorization. Both were company-specific and neither was in the tape beforehand.

The unremarkable volume is the second signal. A 1.0x volume day on a 3.2% move to a record high is not a re-rating; it is an orderly mark-up by holders who were already positioned. There was no evidence of a scramble to establish new positions and no evidence of a sell-the-news exit. That is consistent with a market that already understood the crack environment and treated the print as confirmation.

The divergence between MPC and MPLX on the same day is the third, and it cuts against a lazy reading of the enterprise. Investors marked the refining beat up and the midstream miss down, in the same session, in the correct directions. The consolidated entity is not being valued as a single undifferentiated cash-flow stream.

Street Perspective

Debate: Is this a windfall or a regime?

Bull view: The bull case being made on the Street is that global refining supply was already tightening before the conflict, that demand continues to grow through the decade while net capacity additions are concentrated in petrochemical-oriented facilities, and that the disruption merely pulled forward a structural tightening that was arriving anyway. On this view the current crack environment is an accelerated arrival at a durable destination, and the correct multiple on 2026 earnings is not a trough-cycle one.

Bear view: The bear camp contends that roughly 6% of global refined product capacity is offline for reasons that have nothing to do with the supply-demand balance and everything to do with a conflict that could end with a headline. When it returns, product inventories rebuild, cracks compress toward the pre-conflict curve, and the entire year-over-year earnings step-up unwinds. Capitalizing a war premium at any multiple is the error.

Our take: Both are partly right and the bull case is doing more work than it can support. The structural argument for tight refining is real and predates the conflict; management has been making it consistently for longer than the conflict has existed. But the observable quarter-on-quarter delta is the outage, and management sized it themselves at 6% of global capacity. A framework that treats the structural argument as the base case and the outage as a bonus is defensible. A framework that annualizes Q1 or Q2 2026 into a run rate is not. We think the market is currently doing something between the two, which is why we can respect the business and decline the price.

Debate: Is the buyback accretive or is it top-ticking?

Bull view: Some sell-side desks argue the arithmetic is straightforward and favourable: $8.6B of pro-forma authorization against roughly $76B of market capitalization, funded from a cash flow stream whose capital needs and dividend are covered by midstream distributions. The share count has already fallen 5.8% year over year. Continued repurchase at any reasonable price compounds per-share earnings power, and waiting for a better entry is a market call the company should not be making.

Bear view: The bear camp contends that repurchasing shares at a record price, using cash generated by a temporary margin dislocation, converts a cyclical windfall into a permanent capital loss if the cycle turns. Refiners have a long history of buying most aggressively at the top, precisely because that is when cash is abundant. The absence of any stated pacing framework removes the only defence against that pattern.

Our take: The bear has the better of this one, narrowly, and it is a governance point rather than an arithmetic one. We are not asking management to time the market. We are asking whether repurchase intensity varies with the cycle at all, and the answer given on the call was that discretion is retained without describing how it is exercised. A company that buys ratably through the cycle should say so, because ratable is itself a disciplined framework. A company that says nothing is asking to be trusted at exactly the point in the cycle where trust is most expensive.

Debate: Does the MPLX structure deserve a premium or a discount?

Bull view: A growing consensus view is that MPLX is what makes MPC structurally different from a pure-play refiner. Midstream contributed 53.0% of segment EBITDA even in a quarter when refining tripled, distributions to the parent cover the entire standalone capital programme and the dividend, and the distribution commitment of 12.5% annual growth for two years gives visibility no refining asset can offer. That combination justifies a lower cost of equity than the refining comparables.

Bear view: The bear camp contends that the consolidated financials obscure more than they reveal. Consolidated debt of $32,825M includes $25,634M at MPLX, and noncontrolling interests took $340M of net income in the quarter, so consolidated multiples flatter the parent. Midstream EBITDA fell 7.1% year over year while committing to 12.5% distribution growth, which draws on coverage rather than growth. The market voted on this directly by selling MPLX 2.6% on the day it bought MPC 3.2%.

Our take: The structure is a genuine advantage and the disclosure around it is genuinely awkward. Both things are true. The advantage is that MPC's dividend and capital programme are not hostage to the crack spread, which is exactly why the refining cash flow can be returned rather than retained. The awkwardness is that no consolidated multiple computed off this income statement means very much: enterprise value of roughly $107B against FY2025 adjusted EBITDA of $11,956M gives about 8.9x, and that number includes MPLX's debt and the public unitholders' claim on MPLX's EBITDA. Investors should build the parent's economics from the distribution stream, not from the consolidated multiple. We would like to see MPC publish an MPC-standalone cash flow bridge; the absence of one is the single biggest gap in an otherwise improving disclosure package.

Model Implications

Aardvark Labs does not yet maintain a published financial model for MPC. Rather than present model deltas against a model that does not exist, the table below sets out the drivers any MPC model must carry and where this quarter moved them, so the next quarter can be graded against a stated starting point.

Valuation framework: We are not publishing a price target with this initiation. A target on MPC would require a defensible mid-cycle refining margin assumption, and the honest position one day after this print is that we do not have one that is independent of the conflict. What we can state is the arithmetic: at the $260.51 close, 293M shares outstanding gives roughly $76.3B of market capitalization, and consolidated enterprise value of approximately $107B sits at about 8.9x FY2025 adjusted EBITDA of $11,956M. That multiple consolidates MPLX's $25,634M of debt and the noncontrolling interests' share of EBITDA, so it is not a clean parent multiple and should not be compared directly to pure-play refiners.

What would move us to Outperform: Two of the following three. First, a second consecutive quarter of regional per-barrel EBITDA leadership with the Mid-Continent recovering above the FY2025 R&M average of $5.63 per barrel. Second, a stated repurchase framework that links pace to valuation or cycle position. Third, evidence that the crude and logistics advantage persists at narrower light-heavy differentials, which would establish it as a capability rather than a differential trade.

What would move us to Underperform: A return of disrupted capacity that compresses the R&M margin toward the FY2025 average of $16.87 per barrel, coincident with continued repurchase at levels near the current price and no corresponding step-down in the pace. That combination converts a cyclical peak into a permanent per-share capital loss.

Thesis Scorecard Post-Earnings

This is our initiation on MPC, so there is no standing thesis to grade. The table below establishes the pillars we will carry forward and score every quarter, and marks where each stands as of this print.

Overall: Thesis established. Three bull pillars, two of which are confirmed by this quarter's disclosure and one of which is structurally intact but not yet demonstrated in trajectory. Two emerging risks, both of which are about how the current cycle position is being converted into permanent capital decisions rather than about operating execution.

Action: Hold. We would own this business at a materially lower entry multiple or after one more quarter of evidence that the crude and logistics advantage persists independently of the outage. We would not chase a record close on a quarter whose margin input management itself attributed to a conflict with no timeline.

Bottom Line

MPC delivered the best-executed quarter this management team has put on the board. It ran harder than it guided, did more maintenance than it planned, kept capture at 99% through a violently volatile March, published a new regional disclosure at the granularity where its own performance is weakest, and returned over a billion dollars while doing it. Every one of those is a decision, not a windfall, and each one is checkable against the release.

The problem is not the quarter. It is the price and the posture. The stock entered the print at its highest close of the past year and left the session at a new one, up 55% in four months, on earnings whose entire year-over-year improvement traces to a margin environment management explicitly attributed to roughly 6% of global refined product capacity being offline for geopolitical reasons. Into that, the company announced a $5B incremental repurchase authorization and, when asked plainly how it would pace it, described a philosophy instead of a method.

We are initiating at Hold. That is a statement about the entry, not about the operator. The second quarter should be sequentially much stronger on the controllables alone: roughly 94% utilization, $230M less turnaround expense, and $0.58 per barrel lower refining operating cost. If that materializes alongside a Mid-Continent recovery and any framework at all around repurchase pacing, this becomes a straightforward upgrade. Until then, the fastest-moving variable in the earnings model is a conflict, and neither we nor management is in a position to underwrite it.