Strait of Hormuz Oil Flows: Volumes and Bypasses

Before 2026 the Strait of Hormuz, the channel between Oman and Iran that connects the Persian Gulf to the Arabian Sea, carried about 20 million barrels a day of crude oil, condensate and petroleum products, roughly a fifth of what the world consumes. Pipelines could move around 4.7 million barrels a day around it. The EIA estimates that crude and liquids transiting the strait averaged 4.9 million barrels a day in the second quarter of 2026, down from 21.6 million in the last quarter of 2025, after military action began on February 28.

Most of the effect on the United States came through price. US crude imports from Persian Gulf countries were already down to 490 thousand barrels a day in 2025, under 8 percent of the total, while Brent and the fuel prices that follow it moved with the world market. The EIA’s own tables show that the loss was uneven: Saudi Arabia accounted for 57 percent of the August shut-ins, and by September shut-ins were a fifth of those seven producers’ February output. Every 2026 flow figure here is an EIA estimate with its edition named, and there are no forecasts.

What a normal year looks like

The EIA builds its Hormuz series from Vortexa tanker-tracking data. The most recent full table in its chokepoints report covers 2020 to the first half of 2025, and it shows a flow that was large, stable and tilting from crude toward products.

Crude and condensate transit fell 1.5 million barrels a day from 2022 to the first half of 2025, which the EIA attributes to voluntary OPEC+ cuts starting in 2023, to disruptions around the Bab el-Mandeb that moved some Saudi flow onto the East-West pipeline, and to more refining capacity in Persian Gulf states. Product cargoes rose by 0.5 million. By the numbers in Table 1, the strait handled 4.6% less total oil in the first half of 2025 than at the 2022 peak, but the flow that was left was still, in the EIA’s words, about one-fifth of global petroleum consumption and one-quarter of maritime oil trade.

Where it went is concentrated. The EIA estimates 89 percent of the crude oil and condensate that crossed the strait in the first half of 2025 went to Asian markets, with China, India, Japan and South Korea accounting for 74 percent. A fifth of global LNG trade, 11.4 billion cubic feet a day, also crossed, mainly from Qatar. An interruption at Hormuz is therefore an Asian supply problem first and a global price problem second, which is why the US numbers in the last table are small while the price table is not.

The routes around it, and how much they can carry

The EIA lists three pipelines that avoid the strait. Two of them work, one in Saudi Arabia and one in the UAE, and the third has never carried more than a single cargo.

The arithmetic is simple. At about 20 million barrels a day through the strait and 4.7 million of bypass capacity, pipelines could replace about 23 percent of normal flow. Pipelines also carry oil that already counts in some of the numbers: a barrel from the Persian Gulf that goes by East-West to Yanbu and then through the Bab el-Mandeb does not cross Hormuz, but it does cross another chokepoint, which is why the EIA’s Q2 2026 estimate has Bab el-Mandeb at 8.1 million barrels a day against 5.4 million before the conflict. A closure at Hormuz moves the exposure one strait west, and the August and September 2026 outlooks describe attacks on that route too.

The EIA’s account of 2026, edition by edition

The EIA revises its estimates every month, so the sensible way to read a series like this is by edition. Table 3 lists what each Short-Term Energy Outlook said about the strait in the words of that issue, with its prices. The statements describe the EIA’s understanding at the time, and some were later revised; the last row of the discussion below shows an example.

Note the May shut-in figure. The June edition put it at 11.3 million barrels a day; the October edition, with more tracking data, put the May peak at 10.9. Both are estimates of production that did not happen, and the 0.4 million difference is a reminder that these are not measured volumes. EIA publishes them as estimates in Table 1 of each edition for exactly that reason. For the same reason this page cites the edition each time and avoids a single “current” number.

Who stopped producing, and how much

The EIA’s own table breaks the shut-ins down by country. It starts from each country’s February 2026 crude production and gives the March to July average, August and September. The country detail adds something the headline number cannot: the geography of the shut-ins moved between months. Iraq and Saudi Arabia carried most of the March to July average, Saudi Arabia was most of August, and Iran’s figure rose to 780 and 760 thousand in August and September.

Saudi Arabia accounted for 57% of the August shut-ins and 36% of September’s, and its February production, 10.5 million barrels a day, was the largest in the table. The UAE’s estimated shut-ins were zero by August, which may reflect the Fujairah pipeline in Table 2, although the EIA table does not say so. Total shut-ins in September were 19% of the February base across these seven producers.

What the strait did to prices in the EIA series

Crude price data show the shock cleanly, because Brent, the benchmark for Persian Gulf barrels, jumped on the first trading days of March while WTI, priced in Cushing, moved less. Table 5 lays the monthly averages next to the US retail prices that a driver sees, using the EIA’s monthly series.

From February to April, Brent rose $46.40 a barrel, or 65%, while US regular gasoline rose $1.19 a gallon and diesel $1.78. In dollars per barrel those retail rises are $50 for gasoline and $75 for diesel, against $46 for Brent: diesel rose by more than crude did, which is the distillate tightness described in the diesel guide and the crack spreads guide. The Brent minus WTI gap widened from $6.38 in February to $16.97 in April, narrowed to 59 cents in June and returned to $16.85 in September. The highest daily Brent spot price of 2026 in the EIA series is $138.21 on April 7; the September 15 daily high of $130.80 matches the EIA outlook’s $131. The daily high of $135.51 on October 2 is later than the October outlook’s data cutoff.

The US numbers: small physical exposure, full price exposure

The EIA states that about 0.4 million barrels a day of US crude and condensate imports came from the Persian Gulf through Hormuz in the first half of 2025, about 7 percent of US crude and condensate imports and 2 percent of US petroleum liquids consumption, and that these imports had fallen to the lowest level in 40 years as domestic production rose. The monthly series show how the share moved before and during 2026.

The annual series is a long decline, from about 1.5 million barrels a day in 2015 and 2018 to 490 thousand in 2025, and the monthly series falls off a cliff from April 2026: 775 thousand in February, 186 in May and 113 in June, or 1.9% of imports. Over the same four months total US crude imports fell only 6%, from 6,457 thousand in February to 6,044 in June, while the Persian Gulf portion fell 85%, so other sources made up the difference. The import side of that story, by source country, is in the US imports guide, and the reserve that was drawn alongside is in the Strategic Petroleum Reserve guide.

Ships, insurance and the cost of getting oil out

A strait can be open on a map and closed in practice. The EIA’s March 2026 outlook said the strait was not physically blocked, but that the threat of attack and the cancellation of insurance coverage led most tankers to avoid it. Its Today in Energy analysis of March 26, 2026 reported that tanker rates for very large crude carriers leaving the Middle East for Asia were in March the highest since at least November 2005, when the Argus Freight data begin, and that the high cost of war risk insurance drove crude tanker rates from the Gulf to all destinations to record highs. Loaded vessels confined in the Persian Gulf reduced global tanker availability, so rates from the Americas, especially the US Gulf Coast, also reached records. Clean tanker and gas carrier rates rose as well.

Dates differ slightly even between EIA products. The March 26 analysis describes Iran’s closure of the strait as of March 2, while the Short-Term Energy Outlook describes the strait as effectively closed from February 28, when the military action began. Both can be right: one dates the formal announcement, the other the change in shipping behaviour. A reader building a timeline should say which one is meant, because the volume estimates are quarterly and the price series are daily.

Beyond crude: LNG and propane

Hormuz is also a gas and gas-liquids route. The EIA puts LNG through the strait at 11.4 billion cubic feet a day in the first half of 2025, over a fifth of world trade and mostly Qatari. The EIA’s 2026 outlooks record what followed, again as dated observations.

Both rows show the same structure. The United States produces more gas and propane than its export terminals can load, so the world price moved and the domestic price largely did not: the EIA’s March outlook expected US natural gas prices to be relatively unaffected, and the propane spread widened because high US inventories pushed Gulf Coast prices down while Asian prices rose. The wider the spread, the larger the incentive to export, but the loading capacity, not the price, limits how much can go. That is the same reason the Henry Hub price on the natural gas price page is not the price to read when a story is about Asian or European gas.

Transits, shut-ins and prices measure different things

News about Hormuz quotes at least three different measurements, and mixing them produces apparent contradictions.

Transits are what ships carry

The EIA’s Hormuz flow estimates use tanker-tracking data and cover crude, condensate and products. The EIA outlooks warn that these estimates are hard to make in real time, and press reports of single-week volumes can differ from a quarterly average by a large margin without either being wrong. The Q2 2026 estimate of 4.9 million barrels a day is an average over a quarter that included both near-closure and recovering traffic.

Shut-ins are what producers hold back

A producer that cannot export fills its storage and then stops pumping. Shut-ins lag transits in both directions. Storage can absorb weeks of lost exports, and a restart takes time as well, so a month in which more ships moved can still be a month in which shut-ins stayed high. Table 4 is the shut-in measure.

Prices are what buyers expect

Brent can move on a tanker attack that moves no oil, as the $131 daily high shows, or fall below $70 on an agreement before any volume changes. This is why Table 5 and Table 1 should not be mapped one-to-one. Prices respond to expected flows and stocks, as the STEO guide describes, and the OPEC producers’ capacity to replace barrels is a separate question, treated in the OPEC spare capacity guide. The other chokepoints are on the trade routes map, and none of this section is a forecast.

Data and sources: EIA, World Oil Transit Chokepoints (Table 3 of Hormuz flows 2020 to the first half of 2025; destinations; bypass capacity of about 4.7 million b/d; United States imports from Persian Gulf countries in the first half of 2025; Suez, SUMED and Bab el-Mandeb flows), checked October 23, 2026. EIA Today in Energy, Amid regional conflict, the Strait of Hormuz remains critical oil chokepoint (June 16, 2025) and The Strait of Hormuz is the world’s most important oil transit chokepoint (November 21, 2023: East-West and Fujairah pipeline capacities, Goreh-Jask, unused capacity), checked October 23, 2026. EIA, Country Analysis Brief: United Arab Emirates (July 2026), checked October 23, 2026. EIA, Short-Term Energy Outlook, archived editions of March, May, June, July, August and September 2026, and the October 6, 2026 edition, checked October 23, 2026. EIA monthly series from the petroleum bulk data file updated October 22, 2026: RBRTE, RWTC, EMM_EPMR_PTE_NUS_DPG, EMD_EPD2D_PTE_NUS_DPG, MCRIMUSPG2 and MCRIMUS2, and the daily Brent series for the 2026 high, checked October 23, 2026. Bypass shares, products shares and the price conversions are this page’s arithmetic.