HSTHORMUZ STRAITTRACKER

Analysis

The Hormuz Deficit is All But Gone (for now)

An 11 by 7 grid of recent true-color satellite captures of Gulf and Red Sea loading terminals, with machine-vision tanker detection boxes

I’ve spent the last several weeks analyzing Sentinel 1 and 2 satellite imagery for every major oil loading terminal in the Persian Gulf and Red Sea over the last two years. This analysis comprises 35,000 satellite images of 75 different loading terminals in the region representing more than 95% of pre-war export volumes. I trained a machine vision model to identify and measure tankers on both SAR (satellite aperture radar) and true color satellite images of these loading berths.

Then I measured the level of loading activity on every satellite pass since January 2025. I’ve compared current loading activity visible on satellite with the 2025 baseline to get a “percent of normal loadings” series. Looking at the crisis through this lens, while not perfect, tells a much clearer story than trying to count tankers as they transit a warzone.

Problems with tracking Hormuz itself

In the ~3 months I’ve been counting or attempting to count Hormuz flows, I’ve continually uncovered new patterns of evasion that have surprised me. My initial idea was to count tankers leaving the Gulf of Oman (and then work back from that to an implied Hormuz flow) because this is far enough outside the warzone that there’s little motivation to sail dark.

While this was a good idea that caught significant flows others were missing, it was still fundamentally flawed: as it turns out there are many tankers that sail in and out of the Gulf of Oman and the Persian Gulf dark or near completely dark.

I believe I’ve now done as good of a job accounting for this as anyone can, adding more monitoring areas further downstream. But it remains the case that if a tanker almost never turns on AIS my and most every other tracking method will never count it.

This problem applies even to methods that use private satellite imagery to track vessel movements: in most cases, you’re still only getting 1 pass a day, maybe 2. A tanker sailing across Hormuz can make the trip in just a few hours and they are doing this at night in clustered convoys under US military escort.

The darkness of night means you need SAR to see these convoys, and you need to be lucky enough for the satellite to pass during a very short window or you will miss millions of barrels transiting. Firms like Kpler account for this by making revisions later based on AIS evidence, but this has the same flaws outlined above.

The solution: go upstream and watch the source of flows

Compare all those issues with the simple fact that it is impossible for a tanker loading at a berth to hide itself from the satellites passing overhead. It takes over 24 hours to load most tankers and the tanker has to physically be in position at a known loading berth the entire time.

This means satellite imagery of loading activity is a much more reliable indicator of what’s truly flowing.

There are problems here too: namely, you can’t assume every barrel being loaded into a tanker will make it out through the strait. You could, for example, have one off spikes in loading activity that just flow into oil on water inside the Persian Gulf.

But since the mass exodus of oil on water in June, there has been no correlation between increased observed loadings and increased oil on water inside the Gulf.

Kpler put oil on water inside the Gulf at 60 million barrels at the end of July. This has actually decreased slightly since then to 56 million barrels, but during that same time Persian Gulf loadings have steadily risen. From this we can infer that the oil being loaded is exiting the strait to world markets because it has nowhere else to go.

And conceptually, at this point in the war if there is a tanker loading inside the Persian Gulf it’s a safe assumption that it transited into the Gulf since the war started. It’s difficult to think of a scenario where vessels are steadily flowing in and yet somehow cannot leave.

What the satellite imagery is saying

Below I’ve included the first chart most relevant for this analysis. It shows the percent of normal loading activity observed on satellite for the Persian Gulf since January 2026.

The Hormuz Deficit is All But Gone (for now) | Payload Website Template