For forty years the world built a way around the Strait of Hormuz. In September somebody set the way around alight – and every oil forecast on the street now rests on a pipeline nobody is guarding.
IN BRIEF
- 20 million barrels a day passed through the Strait of Hormuz in 2025. Everything the world built to go around it carries 3.5 to 5.5 million – and both bypass routes have been hit this year.
- Saudi crude supply fell to 6 million barrels a day in August, the lowest in more than thirty years, while world demand is falling 2.5 million barrels a day. Demand is collapsing and the price is still climbing.
- Every 2027 forecast rests on a pipeline assumption. HSBC’s $85 needs bypass flows to reach 6.8 million barrels a day by mid-2027; the EIA’s $74 was modelled on inputs frozen a week before Petroline burned.
On the morning of 10 September, a European weather satellite passing over Saudi Arabia picked up a plume of black smoke drifting across empty desert between Medina and Mahd adh-Dhahab.
There is no city out there. There is a pipe. For the past six months it has been the most important pipeline on earth.
Drones hit its pumping stations in the Riyadh and Medina regions. The Saudi foreign ministry said they came from Iraq. Riyadh shut the line – the East-West pipeline, or Petroline – as a precaution, and Brent broke back above $100 to close the week more than 8 per cent higher.
The reflex is to file this as an oil price story. It is not. It is a story about the difference between having a second exit and having a defended one.
A quarter of a plan
For four decades the world’s oil anxiety had a single address. Around 20 million barrels a day of crude and products moved through the Strait of Hormuz in 2025, roughly a quarter of all seaborne oil trade. Nobody pretended that was comfortable. So the Gulf built around it. Petroline runs about 1,200 kilometres from Abqaiq on the eastern coast to Yanbu on the Red Sea. ADCOP carries Abu Dhabi’s crude to Fujairah, outside the strait.
That was Plan B. It has been under fire all year.
Start with the arithmetic, which was never as reassuring as the engineering. The IEA puts the combined available capacity of the two bypass routes at 3.5 to 5.5 million barrels a day. Against twenty. The detour was never a substitute for the strait. It was a quarter of one, on a good day.
A quarter of a plan
The gold band is everything the world built to avoid the Strait of Hormuz. The hatched area is the traffic it cannot take. Both pipelines have been struck this year.
Bypass capacity 3.5–5.5 mb/d · Hormuz traffic 2025, 20 mb/d
Everything built to bypass the strait carries about a quarter of what the strait carried. Tap a date to see what was hit.
Then count the good days. In March, Iranian drones set fire to the oil storage tanks at Fujairah, the UAE’s exit on the safe side of the strait. In April they hit Petroline itself, knocking out roughly 700,000 barrels a day. In July the Houthis announced that Saudi ships in the Red Sea were now targets, and have spent the weeks since taking the coastline at the far end of that route – first the port of Mokha, then Perim, an island sitting in the middle of the lane every Red Sea barrel has to sail through. Saudi Arabia’s crude supply fell to 6 million barrels a day in August, down 2.3 million in a single month and the lowest in more than thirty years.
Demand is collapsing. The price is climbing.
The result is a market that has stopped behaving the way the textbook promises. Global observed inventories drew 95 million barrels in August, taking the total since February to 507 million. High prices are supposed to cure high prices, and they are dutifully curing demand: the IEA now expects world consumption to fall 2.5 million barrels a day this year, the steepest annual drop since the pandemic, having forecast a 1.6 million decline only a month earlier.
Demand is collapsing and the price is still climbing.
That combination only appears when the shortage is not of oil. There is plenty of oil. What is scarce is oil that can reliably be moved.
Sources: IEA, Strait of Hormuz and Oil Market Report, September 2026.
Every forecast has a pipeline in it
Which is why the forecasts published this month are worth reading upside down.
Every one of them has a pipeline buried inside it. The $74 the EIA expects for 2027 rests on model inputs frozen on 3 September, a week before Petroline burned. The $85 HSBC now carries for next year rests on bypass flows climbing from just over four million barrels a day to 6.8 million by mid-2027. Goldman’s base case rests on the Gulf ending 2027 only half a million barrels short – and the same note concedes $120 if the shortfall turns out to be four million.
Nobody is forecasting oil. They are forecasting pipe, and reporting the answer in dollars.
The forecast is a pipeline bet
Move the dial to set how much Gulf crude reaches the market in 2027 against pre-war levels. The price follows. Every bank number published this month is a position on this one line.
Where the street sits on 2027
The dial interpolates between Goldman Sachs’s three published 2027 scenarios. Sources: Goldman Sachs, HSBC and Bank of America notes, September 2026; EIA Short-Term Energy Outlook, September 2026.
And pipe does not give you a tidy bell around $90. The benign case requires every bypass route to run uninterrupted for twenty-one months, in a war where those routes have already been hit three times. The severe case requires one more drone.
Our reading is that the ceiling is the honest number and the floor has quietly moved. Even if the shooting stops on schedule, oil does not return to the $56 the EIA was forecasting for this year back in January, because what has been repriced is not extraction. It is escort, insurance, freight, and the duplicate infrastructure every importing nation is now discovering it should have funded a decade ago. The risk premium has migrated from the wellhead to the route. Routes do not get cheaper when a ceasefire is signed. They get cheaper when they get boring.
Watch the refined barrel for confirmation. Atlantic Basin refining margins hit record levels in August on diesel cracks, and HSBC has raised its margin assumptions through 2028, expecting product tightness to outlast the crude squeeze. The cost is settling into the product, where households meet it without a benchmark to blame.
India gets the bill in several currencies
India takes this faster than most, and through more doors than one. It imports around 85 per cent of the crude it burns, and a long-standing rule of thumb puts every sustained $10 on the barrel at roughly 0.4 per cent of GDP on the current account. On 10 September, as Brent crossed $100, the rupee slipped to 95.33 to the dollar.
The barrel is the smallest part of that. India’s exposure is denominated in freight rates, war-risk premiums and a currency that softens every time a pumping station catches fire - all of it priced off shipping lanes it neither owns nor patrols. A country can hedge the price of oil. It cannot hedge the Bab el-Mandeb.
Energy security was designed as a map problem. Find another way out, draw it, fund it, build it. 2026 has quietly reclassified it as a defence problem, and no one has budgeted for that.
An alternative route is only an alternative until it is worth hitting.