NYMEX ultra-low-sulphur diesel (ULSD) reaches $4.40/gal on the December 2026 contract by December 31, 2026 in the base case, $4.75 in the bull case and $3.40 in the bear case, on a US distillate stock position 9.5% below last year, refineries pinned at 97.4% utilisation, and exports running 28.5% above 2025.

The base case sits 11.8% above the December 2026 ULSD contract, which settled at $3.9347/gal in the CNBC quote feed read on August 29, 2026, and 4.4% above the Gulf Coast physical spot of $4.216/gal (Energy Information Administration daily spot prices, observation date August 25, 2026). The gap this call trades is not between diesel and crude. It is between the futures curve and the forecast the US government published on August 11, 2026.

Key Levels:

• Instrument: NYMEX ULSD December 2026 contract at $3.9347/gal; Gulf Coast physical ULSD at $4.216/gal — CNBC quote feed, August 29, 2026, and EIA daily spot, observation August 25, 2026
• Base case target: $4.40/gal by December 31, 2026 — above the December forward by 11.8%, above Gulf Coast spot by 4.4%, and 1.3% above the September front month at $4.3452
• Bull case target: $4.75/gal — if distillate stocks print below 95 million barrels before November 1, 2026
• Bear case target: $3.40/gal — below spot by 19.4% and below the December forward by 13.6%; if net exports fall to the EIA’s fourth-quarter assumption of 1.24 million b/d
• Major support: $3.80/gal — the July 2026 monthly average for New York Harbor No. 2 heating oil of $3.803, EIA monthly spot series
• Major resistance: $4.84/gal — 52-week high, CNBC contract data, August 29, 2026
• Invalidation level: two consecutive weekly distillate builds above 2 million barrels each in the EIA Weekly Petroleum Status Report before October 15, 2026

Methodology, and exactly what is being called

Every price here comes from one of three primary sources. Physical spot prices and the weekly balance come from the EIA: the daily spot table released August 26, 2026 for observation date August 25, and the Weekly Petroleum Status Report for the week ending August 21, 2026. Forward prices come from the CNBC quote feed, read on August 29, 2026, with the quote timestamp checked against the session date. Forecasts come from the EIA Short-Term Energy Outlook published August 11, 2026, whose modelling closed August 6. The call is on the NYMEX ULSD December 2026 contract, not the physical spot and not a crack quoted in isolation; cracks are computed as dollars per gallon multiplied by 42, minus the matching WTI contract. Two caveats: August 29 is a Saturday, so forwards reflect the last settlement, and a single Gulf Coast assessment is a thin regional read on a national balance.

The distillate draw is an export story, not a demand story

Distillate stocks fell 2.228 million barrels in the week to August 21, 2026, to 103.391 million barrels — 9.5% below the same week of 2025 and 16.0% below 2024. Ultra-low-sulphur grades alone, the specification that clears into road diesel, fell to 93.590 million barrels, down 11.0% year on year. The draw did not happen because refiners were idle: crude runs were 17.393 million b/d, utilisation was 97.4% against 94.6% a year earlier, and four-week-average distillate production of 5.218 million b/d was up 0.4%.

Domestic demand was not the culprit either. Distillate product supplied ran 3.798 million b/d on a four-week average, 2.2% below last year. The barrels left through the export dock. Distillate exports averaged 1.803 million b/d over four weeks against 1.403 million a year ago — 400,000 b/d more, or 2.8 million barrels a week. That single line explains the entire weekly draw, and it is why the stock number keeps falling into a soft domestic market.

Metric	Week to Aug 21, 2026	Week to Aug 22, 2025	Change
Distillate stocks (m bbl)	103.391	114.242	−9.5%
ULSD, 15 ppm and under (m bbl)	93.590	105.162	−11.0%
Refinery utilisation	97.4%	94.6%	+2.8 points
Operable refining capacity (000 b/d)	18,027	18,160	−133
Distillate production, 4-week avg (000 b/d)	5,218	5,197	+0.4%
Product supplied, 4-week avg (000 b/d)	3,798	3,882	−2.2%
Distillate exports, 4-week avg (000 b/d)	1,803	1,403	+28.5%
Days of cover vs total disappearance	18.5	21.6	−14.6%

Sources: EIA Weekly Petroleum Status Report, tables 1, 4 and 9, week ending August 21, 2026, released August 26, 2026. Days of cover computed as end-week stocks divided by the sum of four-week-average product supplied and four-week-average exports; author’s calculation, not an EIA-published series.

Capacity attrition sits underneath it. US operable refining capacity was 18.027 million b/d, down 299,000 b/d from August 2024, with Gulf Coast utilisation at 97.0% and Midwest utilisation at 101.8% — above nameplate. A system 1.6% smaller and running 2.8 percentage points harder than last summer has already spent the buffer that normally absorbs an autumn turnaround season.

“Strong planning, commercial, and operational execution enabled safe and reliable operations to meet resilient consumer demand. Our results reflect the differentiated capabilities of our value chains and the execution of our optimization strategies.”

— Maryann Mannen, Chairman, President and Chief Executive Officer, Marathon Petroleum Corp. (Second-quarter 2026 earnings release, filed with the SEC on August 4, 2026)

Marathon’s Refining & Marketing margin was $36.33 per barrel in the second quarter of 2026 against $17.58 a year earlier, a result the company attributed “primarily to higher crack spreads in all regions”. Valero reported refining operating income of $4.5 billion for the same quarter against $1.3 billion in 2025.

Why the December contract is mispriced against its own inventory path

The EIA’s August Short-Term Energy Outlook forecasts wholesale diesel at $3.22/gal in the fourth quarter of 2026 and WTI at $74.00/bbl, implying a ULSD–WTI crack of $61.24/bbl. The December futures strip implies $85.28. The forecast requires a $24 collapse in the refining margin that the forward curve does not price at all — and the physical market is already running away from the near end of the same forecast. Gulf Coast ULSD at $4.216/gal is 9.8% above the EIA’s own third-quarter estimate of $3.84, with a month of the quarter still to run.

Measure	EIA August STEO (Aug 11, 2026)	Market or physical (Aug 25–29, 2026)	Gap
Wholesale diesel, 3Q26	$3.84/gal	$4.216/gal (Gulf Coast spot)	+9.8%
Wholesale diesel, 4Q26	$3.22/gal	$3.9347/gal (NYMEX Dec 2026)	+22.2%
WTI, 4Q26	$74.00/bbl	$79.98/bbl (NYMEX Dec 2026)	+8.1%
Implied ULSD–WTI crack, 4Q26	$61.24/bbl	$85.28/bbl	+39.3%
Distillate stocks, end-4Q26	114.1m bbl	103.391m bbl (actual, Aug 21)	−9.4% vs forecast
Wholesale diesel, 2026 average	$3.37/gal	$3.10/gal in the July 7 STEO	revised +8.5% in one month

Sources: EIA Short-Term Energy Outlook, August 2026, Table 2 (energy prices) and the “Notable forecast changes” table; EIA daily spot prices, observation August 25, 2026; CNBC quote feed, read August 29, 2026.

The load-bearing assumption is buried in the balance table. The EIA forecasts distillate stocks ending the fourth quarter of 2026 at 114.1 million barrels against 104.7 million at the end of the third — a 9.4-million-barrel build into the start of heating season, roughly three times the 3.0-million build between the same two quarters of 2025. To get there the forecast needs distillate net exports to fall to 1.24 million b/d, against an actual four-week rate of 1.68 million. Holding every other item in the EIA’s balance constant, an unchanged export rate across a 92-day quarter is roughly 40 million barrels of extra drain. A third of that erases the forecast build.

The steelman for the other side is about crude, not products. WTI averaged $102.13/bbl in May 2026 and $80.46 in July, and the December contract is $79.98; if post-Hormuz normalisation continues, an unchanged crack still delivers a lower flat price. That is why this note calls the outright contract rather than the crack alone: $4.40 at the December WTI price requires a crack near $105/bbl, above today’s spot crack of $93.17.

What the model misses

The framework has three weak points. It treats the export rate as behavioural when part of it is contractual: term barrels committed to Latin American and European buyers do not switch off because a US crack widens, nor do they automatically persist if freight economics turn. Distillate is also the most cyclically sensitive fuel in the barrel, and product supplied is already down 2.2% year on year — a genuine freight recession takes the domestic leg down faster than the export leg holds it up.

The historical analogue cuts both ways. The winter of 2022–23 produced an almost identical setup — distillate cover near record lows, wide cracks, forecasters calling normalisation — and it resolved through demand destruction and mild weather rather than a melt-up. Weather is the largest uncontrolled variable here, and a warm October and November would flatten this thesis without a single inventory number turning.

“Any scenario involving full restoration of inventories, production, and trade flows to pre-conflict levels must account for the partial restructuring of the global oil market that has already occurred.”

— Tristan Abbey, Administrator, US Energy Information Administration (EIA press release, June 9, 2026)

That is the counter-thesis institution speaking, and it sits awkwardly against its own forecast. The agency raised its 2026 wholesale diesel projection by 8.5% in August, yet its August 11 release still expects Brent to “gradually drop to an average of $69/b in 2027”. A forecaster that concedes the market has been structurally rewired, then models a return to pre-conflict inventory behaviour by the fourth quarter, carries a contradiction the futures curve has partially priced and the physical market has not.

What would invalidate this call

The base case to $4.40 breaks if ANY ONE of these four signals fires:

• Two consecutive weekly distillate builds above 2 million barrels each before October 15, 2026. That signals the export rate has broken lower or turnarounds are lighter than assumed — the exact path the EIA’s 9.4-million-barrel build requires.
• Distillate exports fall below 1.45 million b/d on a four-week average. The whole mechanism is the 400,000 b/d export delta against 2025. Remove it and the balance normalises with no change in demand or runs.
• The September 9, 2026 STEO leaves fourth-quarter wholesale diesel at or below $3.30/gal after another month of $4-handle physical prices. If the agency will not mark to market with that much evidence, the gap is a modelling choice rather than a lag, and the convergence trade loses its clock.
• Refinery utilisation holds above 96% through the second half of October. The thesis assumes turnarounds bite into a system with no spare capacity; a light maintenance season restores the buffer and lets stocks build into winter.

What to watch next

The Weekly Petroleum Status Report lands every Wednesday, next on September 2, 2026; the distillate stock line, the ULSD sub-line and the export line are the three numbers that matter, in that order. The next Short-Term Energy Outlook is due September 9, 2026 and is the scheduled event most likely to move the forecast gap — a fourth-quarter revision above $3.50 confirms convergence, a hold at $3.22 extends it. On price, watch for a December close above $4.10, the first real flattening of a curve running from $4.3452 in September to $3.9347 in December.

FAQ

What exactly is the price target on?

The target is $4.40 per gallon on the NYMEX ultra-low-sulphur diesel contract for December 2026 delivery, by December 31, 2026. That contract last traded at $3.9347/gal, so the base case sits 11.8% above the forward. It is also 4.4% above the Gulf Coast physical spot of $4.216/gal and 1.3% above the September 2026 front month at $4.3452.

Why is diesel so far above gasoline right now?

On August 25, 2026 Gulf Coast ULSD settled at $4.216/gal against Gulf Coast conventional gasoline at $3.429 — a 78.7-cent premium. At New York Harbor the gap was 93.2 cents. That is unusual for late August, when gasoline normally holds the seasonal premium, and it reflects a distillate balance 9.5% tighter year on year while gasoline stocks are comfortable.

Is this a bet against crude oil prices?

No. This note takes no directional view on crude; the December 2026 WTI contract at $79.98/bbl is only 8.1% above the EIA’s fourth-quarter forecast. The call is on the product, and specifically the refining margin: $4.40/gal at $79.98 WTI implies a ULSD–WTI crack near $105/bbl, against $85.28 on the December strip and $93.17 in the physical market.

What is the single most important number to track?

Distillate exports in the EIA Weekly Petroleum Status Report. Four-week-average exports of 1.803 million b/d against 1.403 million a year earlier account for 2.8 million barrels a week of extra drain, which is the whole of the recent weekly draw. If that number falls back toward 1.45 million b/d, the thesis loses its engine.

How reliable is the EIA forecast being challenged here?

The EIA is a Tier 1 statistical agency and its weekly data are the industry standard. Its price forecast is a different product. It raised the 2026 wholesale diesel projection from $3.10 to $3.37 between the July 7 and August 11 rounds, an 8.5% revision in one month, and physical Gulf Coast diesel is currently 9.8% above its third-quarter estimate.

Compare the crude leg of this thesis with our Brent to $76 by year-end 2026 half-premium case, the US gas storage analogue in Henry Hub to $2.40 by end-October 2026, the European heating-fuel equivalent in TTF to €85 by end-January 2027, and the currency expression of the same oil path in USD/CAD to 1.43 by Q3 2026.

This article is informational analysis only and is not financial, investment, or trading advice. Foreign-exchange, commodity, and equity markets are highly volatile and can lose substantial value rapidly. Leveraged products carry total-loss risk and may exceed the initial margin posted. Past performance and historical correlations do not guarantee future results. Do your own research and consult a regulated financial adviser before making any investment decision.