The brief: the energy sector is the market's uptrend and Marathon is its best-performing large stock, up about 170% this year on the refining margins the Iran war's product shock produced (a record Gulf Coast crack of $82 a barrel in September against a norm near $20) and a buyback that has retired nearly half the company since 2022. It is ten percent off its high after refiners' worst day since the tariff shock, its options are priced at their one-year high, and it reports on Tuesday. The author is content to own it. This page works through what the cycle is made of, what the filings say the company earns and returns through a cycle, what the market is paying for it versus what it paid in 2022, and which put sits under the levels that matter.
Our view in one paragraph. The cycle is real, the company is the best vehicle for it, and the stock is not cheap. Three facts carry the page. First, 2026 is a product shock, not a crude shock: the Iran war closed Hormuz in late February, and while crude flows were back to pre-war levels by the end of September, Gulf refined-product exports were still running at a fifth to a half of what they were, Russian refinery runs sat at a twenty-year low under Ukrainian drones and a diesel export ban, and a million barrels a day of US and European capacity had closed in 2025–26. Diesel took the shock: the US diesel crack went through $100 a barrel in August, the Gulf Coast 3-2-1 printed an all-time high of $81.86 on 25 September against a long-run norm near $20, Marathon's refining margin went from $16.87 a barrel in 2025 to $36.33 in the second quarter and quarterly earnings from $4 a share to $17.73. The 3-2-1 has since fallen to $41–54 (a G7 stock release, a Russian diesel deal, Chinese exports and crude at $100), still two to three times normal, with the whole excess in diesel and the gasoline crack already back in the teens. Second, the filings show what Marathon does with a cycle: since 2022 it has bought back $36 billion of stock and retired 46% of its shares, the dividend is covered by MPLX distributions alone, and the parent carries $0.4 billion of net debt against $6.7 billion of cash, so an owner at any price is paid 7–8% a year in cash while waiting. Third, the market has already re-rated it: in 2022 the same earnings ($28 a share, 23% free-cash-flow yield) were priced at 4× because everyone expected them to halve, and they did; today's $29 of trailing earnings are priced at 15× and next year's consensus $58 at 7.4×, the top of the company's own ten-year range on every trailing measure. That is a market pricing a plateau, not a peak, and the Street's 2027 estimate already falls 10%. The report on Tuesday is priced at ±9.6% and the beat is not the question (Valero and Phillips 66 already printed $38–39 a barrel for the quarter); the fourth-quarter margin commentary is, after a week in which products fell faster than crude on US–Iran talks and Russian diesel coming back. Our view: sell the December put on Wednesday 4 November with the report known, at the strike the reaction sets, and sell it under the levels rather than near the price: the 18 Dec $390 put at about $12.45 (break-even $378, on the September consolidation and 15% below the price, 25% annualised) if the stock holds $420 through the print; the $380 January put if it does not. An owner at $378 pays 6.5× a down year's earnings for a company that returns 8% of its market value a year, which is the right price to be put the stock at; an owner at $445 is paying for the plateau.
The supply side is documented and unusual. The crude market healed in six months (Hormuz crude flows 13.5 million barrels a day by 30 September, pre-war levels) but the product market did not: Gulf product exports were 677,000 barrels a day against 3.6 million before the war, because damaged hydrocrackers and hydrotreaters are slower to repair than terminals and Gulf refiners kept product at home; Russian runs were 3.8 million barrels a day in June (−30%, the lowest in twenty years) with a diesel export ban from July to September; LyondellBasell Houston and Phillips 66 Los Angeles closed in 2025 (400 kbpd), Valero's Benicia in 2026, and about 900 kbpd west of Suez; China stopped new export contracts in March. Combined Middle East and Russian diesel exports in August were 75% below a year earlier. US refiners ran above 95% from May to September, the highest since 1998, on WTI $8 cheaper than Brent, while simple European and Asian plants earned nothing. Marathon's margin went from $16.87 a barrel in 2025 to $36.33 in the second quarter; Valero printed $39.24 and Phillips 66 $38 for the third, and consensus has Marathon's Tuesday margin near $45. Distillate stocks are 10% below the five-year average (East Coast 32% below, forecast to stay 20–30% below through the winter), and the IEA says middle distillates stay 'tight well into 2027'.
The normalization is also documented, and it started five weeks ago. From the 25 September record the 3-2-1 fell to $41 on 27 October (back to $53 by the 30th): the G7's 100 million barrel release on 2 October, the Trump–Putin diesel deal of 9 October (300,000 tonnes now, up to 3 million later, under a Treasury licence to April 2027; experts call it too small), Chinese diesel exports tripling year on year in September, and Dated Brent up 13% in a month, the crude-cost compression Valero's own COO warned about. The gasoline crack is already in the teens with stocks 2% below normal and demand −0.7% to −2.5%; what is left of the cycle is diesel. The forward curve prices a slow decay, not a collapse: the diesel crack against Brent is $88 for December, $83 for March, $72 for June and $63 for December 2027, still three times the pre-war norm; Goldman's 2027 US diesel margin is $63; Phillips 66 expects today's strength 'through most of next year', HF Sinclair no full recovery 'until sometime in 2028'. Against that, the IEA has 2027 refinery runs recovering by 2.8 million barrels a day while demand recovers by only 2.0, the 2022 template halved cracks within six months and normalized them in two and a half years, and the EPA's record renewable-fuel mandate (with 770 million small-refinery RINs to be reallocated) raises costs for a refiner with no exempt plants. Consensus already has 2027 EPS at $58.44 against $65.07 for 2026.
Verdict: the cycle will not repeat September; the fight is over the slope of the decline. The curve and the Street price a plateau (realized margins of $25–28 a barrel in 2027, double 2025's), the 2023 path prices a slide (EPS $25–35), and the three things that would mark the top are measurable: Gulf product exports recovering toward 3 million barrels a day, the US distillate deficit shrinking toward 5%, and the December-2027 diesel crack breaking $50.
The share count is the clearest fact in the filings: 574 million diluted shares in 2014, 516 million in 2022, 409 million in 2023, 341 million in 2024, 306 million in 2025 and 280.8 million on 30 July 2026. Buybacks were $11.9 billion in 2022, $11.6 billion in 2023, $9.2 billion in 2024 and $3.5 billion in 2025, with $3.25 billion in the first half of 2026 (13 million shares at an average $246) and $6.1 billion of authorization left after a new $5 billion in May. The dividend was raised 10% to $1.10 a quarter in October ($4.40 a year, about 1%); the buyback yield is 6.6%, the shareholder yield 7.6%.
Two qualifications. MPLX, 64% owned, pays Marathon roughly $2.8–3.1 billion a year in distributions (raised 12.5% in October, with 12.5% a year guided for 2026–27), which covers the whole dividend on its own, so the refining cash flow is the buyback. And the buyback's effect shrinks with the price: the second quarter retired 9 million shares at $258; the same $2.5 billion at $430 retires 5.8 million, 2% of the company instead of 3%.
Verdict: the capital-return machine is the reason to own this stock rather than the sector ETF, and it works at every point in the cycle; it just buys fewer shares the higher the price goes, which is one more reason to want to be put the stock lower rather than buy it here.
At 30 June 2026 the parent carried $7.18 billion of debt against $6.74 billion of cash, net debt $0.44 billion, 2% of capital; the other $25.6 billion of consolidated debt sits at MPLX and is non-recourse. Ratings Baa2/BBB, stable. The test case is in the table below: in 2020 Marathon lost $9.8 billion ($15.13 a share), still paid $1.5 billion of dividends, and the stock bottomed at a third of today's book value.
The consolidated statements overstate both the leverage and the cash flow: 100% of MPLX's EBITDA, debt and cash flow are consolidated while 36% of its equity belongs to public unitholders, so the 10% free-cash-flow yield on the table is a few points lower on a Marathon-only basis, and the 9× EV/EBITDA a point or two higher. A sum-of-the-parts (the MPLX stake at market, less standalone net debt) is the honest frame; the direction of the adjustment is the same either way.
Verdict: a would-be owner is not taking balance-sheet risk; the risk is entirely the earnings line and the multiple on it.
On forward numbers it screens cheap: 6.7× 2026 EPS, 7.4× 2027, against a five-year average forward multiple of about 11.7 and an S&P 500 at 22. The sell-side chased the price all October (Mizuho to $457, Barclays $488, Citi $518, TD Cowen $478; the mean target of $424–459 sits at or below the stock), and the 2022 analogue is the bull's favourite: the forward multiple fell to 5.1× in July 2022 and the stock then doubled.
On the company's own history it is at the top of the range: 15.4× trailing earnings against a ten-year mean of 11.8; 9.1× trailing EV/EBITDA against a 2021–25 average of 6.7 and a 2022 low of 3.4; price-to-book 2.9 against 2.0–2.5 in the pre-2022 years. The decisive comparison is 2022 itself: $28.12 of earnings were priced at $116 a share (4.1×, 23% free-cash-flow yield) because the market assumed they would halve, and they did (2024 EPS $10.08). Today's $29 of trailing earnings are priced at $445: the market has re-rated the cycle, not just the earnings. If 2027 earnings land at $40 rather than $58 and the market pays the 9× it paid in 2019 and 2025, the stock is $360.
Verdict: cheap on peak earnings and fully priced on normalized ones; the strike should be chosen for the second description, which is why $390 and not $420.
What confirms it: a close back above the 20-day ($458) and the 27 October gap ($470) on Tuesday or Wednesday; the fourth-quarter margin guided above $30 a barrel; buybacks of $2.5 billion or more in the quarter.
What confirms it: a close below $420 (the 50-day, the put wall and Thursday's low together) after the report; fourth-quarter margin guided under $25 a barrel or the mid-cycle language dropped; buybacks under $2 billion in the quarter.
| Date | Event | Why it matters for the thesis |
|---|---|---|
| Tue 3 Nov (pre-market; call 11:00 NY) | Q3 2026 results | Consensus EPS $25.4 (was $16.08 ninety days ago) and a refining margin near $45 a barrel vs Valero's $39.24; capture vs the $290m of turnarounds; Q4 guide; buyback dollars. The US midterm election is the same day: the diesel-price politics (export-ban threat, Russian deal) ease after it. |
| Wednesdays (10:30 NY) | EIA weekly inventories | Distillate stocks vs the five-year average (−10% on 23 Oct; East Coast −32%): the deficit shrinking toward 5% is a top-of-cycle marker. |
| ~Early Nov | EPA proposal to reallocate the 770m 2025 small-refinery RINs; mid-Nov IEA Oil Market Report | RIN costs for 2026–27 (Marathon has no exempt plants); the IEA's Gulf product-export and 2027 run numbers. |
| Wed 4 Nov | The morning after | The recommended entry: the strike the reaction sets, with the report known and the November premium still in the chain. |
| Fri 6 Nov / Fri 13 Nov | MPLX Q3 distribution record / payable | $1.2111 a unit, +12.5%; about $2.8–3.1bn a year to Marathon. |
| ~16 Nov | Q3 13F filings | Who bought the summer rally; BlackRock and BNY Mellon added in Q2. |
| Wed 18 Nov / 10 Dec | Ex-dividend $1.10 / payment | The raised dividend; $4.40 a year. |
| 20 Nov | November monthly expiry | The report's volatility premium is gone by here; the first expiry after the print. |
| Nov–Dec (rolling) | US–Iran talks; Russian diesel output and sanctions waivers; Venezuelan crude; weekly EIA distillate stocks | The crack-spread tape that moved the stock 7% in a day; winter distillate is the bull's card. |
| Late Nov / early Dec | OPEC+ meeting | Crude supply; refiners prefer cheaper crude with firm products. |
| 9 Dec | FOMC decision (a hike 85% priced) | Rates: a low-beta (0.54) stock, but the sector trade is a rates trade at the margin. |
| 18 Dec | December monthly expiry | The recommended put's expiry; quad witching. |
| 24 Dec | Target date | Every structure on this page is valued on this day. |
| Early Feb 2027 | Q4 2026 results; 2027 capex and Q1 turnaround guide | Outside the January expiry; the next time the mid-cycle thesis is marked. |
Daily closes. Hover for values; the table shows the same data.
| Total return | 1 week | 1 month | 3 months | 6 months | YTD | 1 year |
|---|
Marathon is up this year against for Valero, for Phillips 66 and for the energy ETF; over one month it is against for Valero. The refiners are the sector's leadership and its beta: they rose 55% in the third quarter alone and gave back 7–9% in a day on 27 October. Marathon's edge over Valero this year is the buyback (fewer shares every quarter) and the Mid-Continent exposure; Valero's over Marathon is the cleaner balance sheet and the Gulf Coast export position. Extreme outperformance of this kind persists while estimates rise and reverses hard when they stop; the October target raises say they are still rising.
Ten years of filings tell the cycle in three lines. Earnings: EBITDA ran at $5–8 billion a year from 2014 to 2019, went to −$8.9 billion in 2020, peaked at $24.7 billion in 2022 (EPS $28.12), fell to $17.8 billion and $10.1 billion in 2023–24, steadied at $11.5 billion in 2025 (EPS $13.22) and is $17.4 billion over the twelve months to June 2026 with the record third quarter still to be added; the second quarter alone earned $17.73 a share, more than any full year before 2022. Cash returned: buybacks of $11.9, $11.6 and $9.2 billion in 2022–24 and $3.5 billion in 2025, plus $1.1–1.5 billion of dividends a year, took the diluted share count from 574 million in 2014 and 516 million in 2022 to 292 million over the last twelve months and 281 million today; free cash flow was $13.9 billion in 2022 and $12.9 billion over the trailing year on consolidated numbers (MPLX's cash flow is inside that). Valuation: at the end of 2022 the market paid 4.1× earnings, 3.4× EBITDA and a 23% free-cash-flow yield for the peak year, because it expected the earnings to halve, which they did; at the end of 2025 it paid 12.3× and 7.4×; today it pays 15.4× trailing earnings, 9.1× EBITDA and a 10% free-cash-flow yield, and 7.4× the Street's 2027 EPS of $58. The multiple has gone up with the earnings this time. For a put seller the useful arithmetic is the inverse: at the $378 break-even of the December $390 put the stock is 6.5× 2027 consensus and 9.5× a normalized $40, which is the multiple the market paid in 2019 and 2025 for a mid-cycle Marathon.
EBITDA (operating income plus depreciation), free cash flow (operating cash flow minus capital expenditure) and cash returned to shareholders (buybacks plus dividends), by fiscal year.
P/E and EV/EBITDA on each year's earnings at the year-end price (left); free-cash-flow yield (right). Loss years have no multiple. The last point is the trailing twelve months at today's price.
Revenue growth (bars, left) with gross and operating margins (lines, right). Multiples are paid for the first and defended by the second; the TTM point is the last four quarters.
Diluted weighted shares (left) and diluted EPS (right). A falling share count is the second engine of per-share growth: the same earnings land on fewer shares each time; a rising one (stock compensation, acquisitions) is a tax on it.
Blue candles closed up, red closed down. Orange dashed: dealer-gamma levels (call wall, put wall, gamma flip); swing support and resistance are listed in the key-levels table. Hover for the crosshair: the price is tagged on the price axis and the date on the time axis. Right panel: net dealer gamma per strike on the same price axis (blue where dealers are long gamma, red where short).
Trend: a near-triple in ten months, then the first real pullback. Marathon closed 2025 at $161.93, its 52-week low, and ran to $493.98 on 19 October with only two drawdowns of more than 10% along the way; the 50-day average has been below the price since February and sits at , the 200-day at , 53% below the price, which says how far this move is from anything an average would call home. The last two weeks are the correction: an intraday reversal at the high on 19 October (a $494 print, a $473.69 close), lower highs on 21 and 26 October, then −7.4% on 27 October on 1.4 million shares (one and a half times average) as the sector was sold to fund the chips rally, a probe to $419.27 on 29 October that was bought back to $432.51, and Friday's +2.8% to $444.73 on a full-bodied up candle. Ten percent off the high, RSI , the 20-day average at $458 now above the price for the first time since the summer.
Levels. Above: the 20-day at $458, the 26 October high at $478, then $494. Below: the 50-day at , the dealers' $420 put wall and Thursday's $419 low together, then the builder's support clusters at $379 (the September consolidation) and $348 (the July–August base); the 200-day at $290 is a different market. The report on Tuesday 3 November, before the open, lands inside all of this.
Volume. The three sessions of the 27–29 October slide ran at 1.1–1.4 million shares against a 20-day average of , and the two up days since (26 October, 30 October) at 0.7–1.0 million: the selling was heavier than the buying, which is what a first pullback in a strong trend usually looks like, and not what a top looks like (those come with the heavy volume on the up days first).
Technical verdict: a strong uptrend in its first real pullback, with the 50-day average ($418), the $420 put wall and Thursday's $419 low stacked within $2 of each other, five percent below the price, and the next support twenty percent lower at $379. That is where a put strike belongs: under $420, at or below $390, so that the first test of the trend can fail and the position still be whole. The thesis of a retest of $485–494 needs a close back above the 20-day at $458 first; until then the stock is in a $420–478 range and the report decides which side it leaves it from.
| Level | Price | vs spot |
|---|
Support and resistance are clustered swing lows and highs from the last eight months of daily bars; the gamma levels come from CBOE open interest.
What the profile says for Marathon. A thin book for a stock this size (open interest about contracts, net of gamma per 1% move), so the levels describe where the positioning sits rather than what it can enforce. It is marginally positive around the price: the largest strike is the $430 call (dealers long, +$4 million), the $450 call wall sits one percent above Friday's close and the $420 put wall five percent below, and the flip is far down at $398; between $420 and $450 the dealers lean against moves, below $398 they would start selling with the stock. For a put seller the structure says the obvious thing twice: the $420 strike is where the hedging support is, the $398 flip is where it ends, and a strike at $390 or below sits under both. The sign convention (dealers long customers' sold calls, short customers' bought puts) is an assumption; open interest is end-of-day (see the footnote under the chart), and the report on Tuesday will reshape this chain by Wednesday.
How to read this. Market makers who sell options hedge them with stock; how much stock they must buy or sell as the price moves is their gamma. We estimate it from CBOE open interest for every listed strike and expiry, recompute each option's gamma with Black-Scholes from its own implied volatility, and sign it by the usual convention: dealers are long the calls customers sell and short the puts customers buy. Positive net gamma means dealers sell as the stock rises and buy as it falls, dampening moves; negative means they chase the move and amplify it. The sign convention is an assumption, and open interest is end-of-day, so the profile changes after every expiry.
Realized volatility from daily closes; implied volatility is the one-year series from Interactive Brokers' 30-day implied-volatility history, through Friday 30 October.
Solid: implied volatility of the at-the-money option for each listed expiry, on a calendar axis. Dashed: the forward volatility the market prices for the period between one expiry and the next; the spike is the earnings report. From CBOE's delayed chain.
Last three months of closes, then the range the market prices to each expiry: today's price plus and minus the at-the-money straddle (shaded, the expected absolute move) and ± one standard deviation from at-the-money IV (dotted). Green dashed: the report's target.
Out-of-the-money puts left of spot, calls right of spot, for the first monthly expiry after a month, the expiry that carries the earnings report, and January. A smile that rises to the right means upside calls are bid: the market pays up for a squeeze.
The premium is at its one-year high, and the report is why. Thirty-day implied volatility is , the th percentile of the year (rank ), against realized over the last month. Refiners' options were priced in the thirties for most of the year; the October slide and Tuesday's report have pushed them into the fifties. A seller today is paid for both.
The term structure dates the event. At-the-money volatility is 59% for 20 November, the first monthly after the report, 54% for 18 December, 52% for January and 52% all the way out to September: the curve is flat after November, which means the market expects the stock to stay a 50%-volatility name through the winter, report or no report. The November straddle prices ±11.4% for three weeks, of which the three-day report is perhaps half; the December straddle ±15.6%, January ±18.9%. Marathon's own reports have moved it 3–8% in the last two years; the pricing is generous to the event and the post-event months both.
The smile is flat. Twenty-five-delta skew is near zero in every expiry: puts and calls trade at about the same volatility, which is unusual for a stock ten percent off a parabolic high and says the chain is positioned for a squeeze as much as for a fall. A put seller is not being paid a fear premium; the edge is the absolute level (50%+ for a refiner) and the time.
The cone. The one-standard-deviation move to 24 December is ±, 22% of the price; a $390 strike is 0.6 standard deviations below spot for December, $380 for January 0.7. The market gives a retest of $485 by Christmas a chance.
| Structure | How it is built | What you want | Risk | Best used when |
|---|---|---|---|---|
| Long call | Buy a call | A big move up, fast | The premium (100% if it expires below the strike) | You expect a move larger than the market prices; implied volatility is low |
| Bull call spread (debit) | Buy a call, sell a higher call | The stock at or above the upper strike at expiry | The net debit | You have a price target and a date; you want the highest return per dollar to that target |
| Cash-secured short put (credit) | Sell a put, hold the cash to buy the shares | The stock to stay above the strike; you keep the credit | Owning the shares at the strike in a sell-off (large, stock-like) | You would be happy to own the stock lower; volatility is high |
| Bull put spread (credit) | Sell a put, buy a lower put | The stock above the short strike at expiry; keep the credit | Spread width minus credit (defined) | You are bullish-to-neutral and want to be paid to wait with a known worst case |
| Risk reversal | Sell a put, buy a call (often for zero cost) | A large rise; behaves like stock above the call strike | Stock-like below the put strike; margin required | You want stock exposure without paying premium and accept the downside |
| Put-financed call spread | Sell a put, buy a call spread | The stock at the top of the spread; the put pays for it | Stock-like below the put strike | You want a call spread for free and accept assignment risk |
| Put spread + call spread | Sell a put spread, buy a call spread | The stock above the call spread's top strike | Both spreads' widths minus net credit (defined both ways) | You want a leveraged payout at the target and a known worst case, with part of the cost paid by the put spread |
Returns on the short-put structures below are shown on a cash-secured basis (capital = strike minus credit). Brokers typically require about 20% of that as margin; for a stock this volatile expect more.
The brief is to be paid to own a refiner that has tripled, in a sector at its high, with a report three sessions away. Three expiries are priced: 20 November (the monthly after the report, 20 days), 18 December (48 days) and 15 January (76 days, with the next report in late January just outside it). The target of $485 (a retest of the October high, +9%) and the 24 December date exist so every structure is valued on the same day; for a short put the question is where the strike sits against the 50-day and the $420 wall, the $379 support, the ±11% the report is priced at, and what the premium pays per month of that.
Scenarios on 24 December: Bear $356 (−20%, a margin reset or a sector rotation that takes out the $379 support; implied volatility +5 points) · Flat $445 (volatility −2 points) · Base $480 (+8%, back toward the high) · Bull $485 (the target; volatility −5 points) · Overshoot $520 (+17%, the Street's high targets).
| Trade | Net | Max loss | Break-even | Delta | PoP |
|---|
Net is the mid-price debit (paid) or credit (received) per share; one contract is 100 shares. Max loss is the capital at risk per share (for a short put, the strike minus the credit). PoP is the lognormal probability, at today's implied volatility, of finishing above break-even at expiry. Scenario cells show profit or loss as a percentage of the capital at risk on 24 December (20 November trades have expired by then, so their cells are the result at expiry); −100% means a total loss. Highlighted rows are the recommendation.
| Trade | Capital at risk | P&L | Return |
|---|
Pick up to four trades. Values on the target date use Black-Scholes with each contract's own implied volatility; trades expiring before then show their value at expiry.
The recommendation is about timing first and strike second. Tuesday's report is priced at ±9.6% in a stock whose last four reactions were −6%, +6%, +3% and +2%: the premium is fair, and the thing that moves the stock (the fourth-quarter margin commentary after a week of crack compression) is a coin the peers' prints have not flipped. So: not before the print unless at half size; on Wednesday 4 November, the December put at the strike the reaction sets, under the $420 cluster where the 50-day, the put wall and Thursday's low sit, so that the first failed test of the trend leaves the position whole.
| Strike | Bid | Ask | Mid | Break-even | IV | Delta | Theta/day | Volume |
|---|
Option quotes, Greeks and the one-year implied-volatility series: Interactive Brokers at Friday's close (fetched ); open interest, term structure, smile and dealer gamma: the CBOE delayed chain, open interest of 29 October. Daily prices: Interactive Brokers to 30 October. Financial history: SEC XBRL company facts (10-K and 10-Q filings 2014–2026; EBITDA = operating income + depreciation; free cash flow = operating cash flow − capital expenditure; year-end multiples at the last close of each year), consensus from Yahoo Finance. Company facts and quotes: Marathon Petroleum Q2 2026 results and 10-Q (4 Aug 2026), FY2025 results (3 Feb 2026), the 3 Nov report notice, MPLX distribution release (Oct 2026); Valero and Phillips 66 Q3 2026 releases (22–29 Oct); 24/7 Wall St (23, 28 and 29 Oct 2026), Yahoo Finance/Zacks (29 Oct), MarketBeat (Oct 2026 analyst actions and consensus), Stockanalysis (31 Oct statistics), Trefis (28 Oct), FullRatio and ValueInvesting.io (historical multiples), EPA (RFS 2026–27 final rule; 31 Aug 2026 SRE decisions), S&P Global (California closures, May 2026), Rio Times and Leadership.ng (Dangote exports, 2026), MarketChameleon (earnings-day moves; vendors conflict and are flagged in the research notes). Refining cycle: IEA Oil Market Report (October 2026) and IEA Russian-refining commentary (17 Sep 2026), EIA (Q2 2026 review; weekly inventories to 23 Oct; October STEO), CNBC (30 Sep 2026 Hormuz flows; 17 Aug 2026; 2 Oct 2026), Columbia CGEP ('Beyond Crude'), Kpler (Jan 2026), Stillwater Associates (9 Sep 2026), Crossvol Research (forward curve, 17 Oct 2026), Goldman Sachs via Oilprice (31 Aug 2026), CFR and NPR (Russian diesel deal, 9–10 Oct 2026), Argus (California imports, 27 Apr 2026), thetrading.tools and globaloilshock (3-2-1 history; trackers differ on the gasoline leg), Valero and Phillips 66 Q3 2026 call transcripts (Motley Fool). Research notes: research_notes/Marathon and Valero refiners deep dive/. Written 31 October 2026; the report on 3 November and the February 2027 guide will make the prose stale.