Why this matters for the Americas: Hormuz affects the global marginal barrel, freight and refining economics even where direct Gulf import dependence is lower.
OilWatch Network Investigation — the headline numbers hide several measurement traps. "Severe" does not mean closed. "80 incidents" does not mean 80 attacks. And the traffic numbers cannot simply be divided by one another.
There is a strange contradiction sitting at the centre of the oil market.
Brent was trading at about $84 a barrel on Monday morning, 10 August, with Reuters reporting $83.74 at 0807 GMT — far below the crisis highs, and up from below $80 last week, even as another vessel was struck in the Strait on 8 August. Diplomatic reports suggest Iran and Oman are close to agreeing new shipping arrangements through the Strait of Hormuz. Yet Iran has simultaneously said that an agreement with Oman would not itself reopen it: Tehran says wider conditions involving the United States must also be met.
Then there is what is actually happening at sea.
On 8 August, two days before this article was written, UK Maritime Trade Operations reported another attack 18 nautical miles east of Khasab, Oman. A vessel was struck by an unidentified projectile and caught fire before the blaze was extinguished. The vessel and crew were subsequently reported safe.
The UAE separately said the vessel was connected to ADNOC and accused Iran of carrying out the missile strike. ADNOC says 15 of its vessels have been attacked by missiles or drones since the conflict began, with one crew member killed and 20 injured across those incidents. ADNOC itself did not identify those responsible; the attribution to Iran came from the UAE government.
So which signal is right? The $84 oil price? The diplomatic negotiations? Or the missiles still striking ships?
The answer is that they measure different things.
And that is also the central lesson hidden inside one of the most useful official documents produced during the Hormuz crisis.
On 4 August, the Joint Maritime Information Center published Update 080, covering maritime security from 1 March to 4 August. The 20-page document, hosted by UKMTO, pulls together threat assessments, shipping data, vessel incidents, naval enforcement, mines, navigation interference and maps of recent attacks.
Read quickly, it appears to describe a single picture of the Strait. Read carefully, it describes several different pictures at once. And some of its headline numbers cannot safely be compared with one another.
"Severe" does not mean "closed"
The first distinction is the simplest and perhaps the most important.
JMIC rates the Strait of Hormuz at SEVERE. It describes deliberate hostile action against shipping as "highly likely under current conditions." The Gulf of Oman is one category lower, at SUBSTANTIAL, while the Arabian Gulf is MODERATE.
That is a security assessment. It is not a measurement of how much of the Strait is closed.
A waterway can have some traffic moving through it while simultaneously presenting a severe probability of hostile action to the vessels that do transit.
This sounds obvious. But it becomes important because three separate questions are repeatedly collapsed into one: is the Strait physically passable; how much commercial traffic is actually passing through it; and how dangerous is it to do so?
JMIC's SEVERE classification answers the third question, not the first two.
That distinction matters enormously when markets begin to interpret a handful of transits as evidence that Hormuz has "reopened."
The 80-incident number is not 80 attacks — and they are not 80 Hormuz attacks
Update 080 reports 80 confirmed maritime-security incidents. It would be very easy to turn that into 80 attacks on shipping in the Strait of Hormuz. It is neither.
Two separate compressions happen at once.
The first is about event type. The annex contains different classifications: actual attacks, serious attacks and near misses, but also suspicious activity and other maritime-security events. One April entry is logged as suspicious activity after a vessel experienced aggressive hailing and threats during a Hormuz transit. Another records a splash close to a vessel. Elsewhere the document records projectile strikes that caused fires, structural damage and casualties. These belong in the same security record. They do not represent the same phenomenon.
The second compression is geographic, and it is the one most likely to be missed. JMIC states a "Total of 80 incidents since 01 March," while the annex itself is headed "Confirmed Maritime Security Incidents List since 28 February" — a compilation across JMIC's wider maritime reporting area. That area includes the Strait of Hormuz, the Arabian Gulf and the Gulf of Oman, but it also includes the Gulf of Aden, the Southern Red Sea and elsewhere. The 80 is a report-wide figure. It is not a Hormuz figure.
So "80 attacks in Hormuz" gets both dimensions wrong simultaneously. But dismissing the non-kinetic entries would miss something real too: coercive hailing, surveillance, routing pressure, boarding, naval enforcement and near misses are part of the operating environment even when a missile does not hit a hull.
The right question is therefore not how many incidents. It is: what kind of incidents, in which body of water, attributed to whom, producing what consequence?
So we classified the annex ourselves, row by row, on both dimensions.
Of the 86 vessel rows printed across pages 12 to 20, 52 are kinetic — entries JMIC types as attack or serious attack. The remainder are three near misses, 22 records of suspicious activity, and nine advisories, blockade-enforcement actions, a hijack and an illegal boarding.
Of those 52 kinetic rows, 40 fall definitely within the Strait of Hormuz, the Arabian Gulf or the Gulf of Oman. Seven are elsewhere entirely — the Gulf of Aden, the southern Red Sea, the Arabian Sea. The remaining five cannot be settled: JMIC gives them only as a bearing and distance from "Oman," with no locality or coordinates sufficient to establish which side of the Gulf of Oman / Arabian Sea boundary they fall on. If all five turn out to lie inside the Gulf of Oman, the theatre figure is 45.
So the defensible statement is this: of the 52 kinetic entries in JMIC's annex, 40 — possibly as many as 45 — occurred in the Hormuz theatre. Not 80. And the non-kinetic majority of the annex is not a rounding error; it is 34 of the 86 rows.
The report does not reconcile with its own annex
There is one more thing, and we did not expect to find it.
JMIC states a total of 80 incidents since 1 March. Its annex prints 86 vessel rows representing 85 unique UKMTO report references — one report number, UKMTO 021-26, is carried across two rows for two vessels, the SAFESEA VISHNU and the ZEFYROS. The document does not explain the reconciliation.
There is an obvious-looking answer, and it is worth setting out precisely because we cannot rely on it. Five annex rows are labelled Advisory, which would arithmetically reduce 85 unique report references to 80. But Update 080's increase from the previous report includes 103-26, itself labelled Advisory, so we cannot treat that arithmetic coincidence as the counting rule without inventing one.
We therefore do not know JMIC's counting rule, and we have not guessed it. Every figure above is a count of annex rows, because the annex is what we can actually audit.
That is also why this piece contains no sentence of the form "X of the 80." Pairing our row-level numerator with an aggregate whose composition is not published would commit precisely the error the rest of this article is about.
(Warning 108-26, the 8 August strike east of Khasab, post-dates Update 080's 4 August cut-off and forms no part of these counts.)
The traffic numbers contain an even bigger trap
JMIC gives the historical Hormuz baseline as approximately 138 vessels per day, on a 2025 reference. It then reports 30 U.S.-facilitated Strait of Hormuz transits during 2–3 August, a 48-hour period.
The obvious calculation is tempting. Thirty transits against a normal 276 over two days produces roughly 11% of normal.
Do not make that calculation. JMIC itself, immediately beneath the figures, warns why.
The numbers are drawn from two different datasets covering different populations of vessels. U.S. NCAGS data count U.S.-facilitated Hormuz transits only. S&P Maritime/SeaVision data cover non-facilitated cargo-carrying vessels above 1,000 deadweight tonnes, and exclude vessels calling at Iranian ports adjacent to Hormuz. Those are not interchangeable numerators and denominators.
The annex drives the problem home. Its daily tables show extremely low cargo and tanker counts during the final week covered by the report — in several cases zero, one, two or three vessels in a given category — while immediately below them sits another UKMTO 90-day transit chart constructed from its own reporting universe.
All of these datasets tell us something. They do not necessarily tell us the same thing.
So what percentage of normal traffic can we defend?
Rejecting the JMIC calculation does not mean percentages are impossible. It means the numerator and the denominator have to measure the same thing.
Aggregate traffic counts are already circulating. Reuters reported 33 vessels transiting Hormuz from Monday to Thursday, 3–6 August, against 50 over the same four days a week earlier. It said roughly 130–140 ships would typically transit the Strait before the closure. That is a legitimate measurement, and it is the kind of figure that gets compressed into a single headline percentage.
OilWatch's own series does something different, and narrower. It uses IMF PortWatch's daily chokepoint data: satellite-AIS observations of the same fixed Hormuz geofence, counted with the same vessel-classification field, compared against that same dataset's own 2023 average. Same provider, same population definition, same boundary, same method on both sides of the ratio.
On that like-for-like basis, for the seven days to 2 August 2026:
- AIS-observed tanker transits: about 2% of the 2023 tanker-count baseline.
- Tanker deadweight tonnage: about 3% of its 2023 baseline — reported here as a secondary figure, because at these tiny volumes a single large vessel dominates a week's tonnage. One day in that week accounts for the majority of it. The count is the more stable measure.
The two figures — Reuters' aggregate traffic count and the PortWatch tanker series — should not be treated as competing estimates of the same thing. They do not measure the same vessel population, they do not use the same baseline, and they do not cover the same window. Setting them side by side is useful. Dividing one by the other is not.
A separate PortWatch all-vessel series also shows traffic at a small fraction of its 2023 baseline, but its geography and observation window are not identical to Reuters/Kpler's measure, so the two figures should not be treated as directly comparable.
What makes the PortWatch figure worth stating is not that it is the true number. It is that its numerator and denominator were generated by the same measurement system.
To put the tanker count in human terms: in 2023, PortWatch recorded an average of 48.8 tankers a day through Hormuz — roughly one every half hour. In the seven days to 2 August 2026, it observed seven tankers in total, with two days recording none at all. In 2023, Hormuz moved that many tankers in about three and a half hours.
The limitation of our own better number
There is an important limitation, and it applies to our figure as much as to anyone else's.
PortWatch measures observable AIS traffic. Vessels operating dark are not captured, and dark transits have become materially more important during this crisis — we have reported crossings in which every vessel had its transponder off. The 2023 baseline contains no comparable dark-transit behaviour, so the current and historical sides of the ratio are not equally observable even though they are equally measured.
The ~2% figure should therefore be read as an AIS-observed floor for the week to 2 August, not a complete census of every tanker that physically crossed the Strait. The true number is higher by an unknown margin, and the direction of that error is the only part of it we can state with confidence.
That is a weaker claim than a clean percentage. It is also the only one the data supports.
The "safe Omani side" is another dangerous simplification
There is an increasingly intuitive picture of the Strait in which Iran controls the northern side while vessels can use a safer southern route through Omani waters.
The UKMTO evidence does not justify treating the Omani route as safe.
Update 080 explicitly records confirmed projectile strikes and near misses inside Omani territorial waters and on the approaches to Hormuz, and says its geographical focus remains the Limah–Khasab axis.
The incident map makes this visually obvious. Recent events are heavily concentrated around the narrow entrance to the Strait and the Musandam peninsula rather than simply hugging the Iranian shore.
The report's most recent incidents before publication included vessels struck or threatened while operating in Omani waters. And four days after Update 080, UKMTO Warning 108-26 recorded another projectile strike just 18 nautical miles east of Khasab.
That does not mean Omani territorial waters and Iranian-controlled waters carry identical risks. It means something more precise: "Omani corridor" must not become shorthand for "safe corridor." The physical incident record does not support it.
This is no longer simply Iran versus commercial shipping
There is another part of the report that deserves more attention than it has received.
Commercial ships are operating between two different systems of military control.
On one side is Iran. JMIC describes continued IRGC monitoring, hailing, surveillance and routing pressure. It says AIS-equipped vessels transiting Hormuz should expect directed hails telling them to move towards the northern Iranian-controlled route.
On the other is the United States. JMIC says CENTCOM reported that, as of 3 August, 44 commercial vessels had been redirected, two disabled and two boarded for compliance verification as part of U.S. blockade enforcement around Iranian ports.
JMIC warns that unauthorised entry or departure from blocked Iranian ports may result in interception or force. Neutral transits remain permitted, but vessels are told to expect verification queries, visit-and-search activity and a strong U.S. naval presence. It recommends keeping 30 nautical miles from U.S. units to reduce misidentification risk.
So the operating environment is not adequately described as Iran attacks vessels → shipping avoids Iran. It is closer to: Iranian coercion and kinetic risk, plus U.S. blockade enforcement, plus mine risk, plus electronic interference, plus severely reduced commercial traffic.
That is a much more complicated maritime system.
The AIS paradox
Then comes perhaps the strangest operational contradiction in the report.
AIS — the Automatic Identification System — exists in part to make vessels visible. In normal navigation that visibility is a safety feature. In a contested Strait, visibility can become something else.
JMIC says vessels are generally expected to maintain AIS transmissions unless security conditions prevent them from doing so. But the same document warns that vessels transiting Hormuz with AIS active can expect directed Iranian hails and routing instructions. It separately says vessels remaining stationary for extended periods with AIS on may face elevated targeting risk.
That creates an uncomfortable dilemma. Being visible helps other ships and military authorities identify you. Being visible can also make it easier for a hostile or coercive actor to locate, identify and contact you.
The report does not advise vessels simply to turn AIS off. And an April incident recorded in the annex demonstrates why the problem is not that simple: a container vessel attacked off Oman had AIS switched off and nevertheless came under RPG and small-arms fire.
Visibility is therefore neither wholly protective nor wholly dangerous. It has itself become a risk variable.
Even the measuring system is being disrupted
There is a final layer to the measurement problem.
Much of what the outside world knows about shipping disruption comes from electronic vessel-positioning and tracking systems.
But JMIC says GNSS interference remains present across the Hormuz approaches, the Gulf of Oman and the Arabian Gulf. Across other parts of its Voluntary Reporting Area, particularly around the southern Red Sea, it also reports broad-spectrum GPS/GNSS spoofing and false AIS manipulation. Its own anomaly maps show substantial concentrations of geographically impossible AIS positions.
This does not make vessel-tracking data useless. It does introduce another uncertainty that should accompany apparently precise traffic numbers.
The system being used to observe the disruption is itself operating in an electronically disrupted environment.
That is an uncomfortable fact for anyone attempting to produce a single definitive real-time count.
And then there is the oil price
Against all of this, Brent was sitting close to $84 a barrel on Monday morning.
That price matters. It tells us the market is assigning less value to the worst-case Hormuz scenario than it did at previous moments in the crisis. Diplomacy matters. Expectations of future reopening matter. Alternative routes, inventories, changed demand expectations and the duration investors expect the disruption to last all matter to price.
But Brent is not a maritime-security sensor.
A falling oil price does not tell us that projectile strikes have stopped. A rising oil price would not prove that the Strait had physically closed.
Price is a market manifestation of the crisis — one output of a much larger system. The physical mechanism can remain active while its market manifestation recedes.
That appears to be precisely what is happening now. Iran says its shipping agreement with Oman is nearing completion, but also says that agreement by itself will not reopen Hormuz. Meanwhile vessels are still being attacked.
What UKMTO's report actually tells us
The strongest conclusion from Update 080 is therefore not that the Strait of Hormuz is open or closed. It is that normality has not returned.
Commercial traffic remains deeply depressed. JMIC judges deliberate hostile action against ships to be highly likely. Projectile attacks have occurred in Omani waters as well as around the Iranian-controlled approaches. Mines remain a reported hazard. Iranian forces are monitoring and directing shipping. U.S. forces are simultaneously enforcing a blockade. And navigational interference complicates both vessel operations and the outside world's attempt to measure them.
Some vessels are getting through. That is not the same thing as the Strait reopening.
And it is precisely why the numbers need to be handled so carefully.
SEVERE is not a measure of closure.
80 incidents are not 80 attacks, and they are not 80 Hormuz incidents. Fifty-two of the annex rows are kinetic; 40 of those are in the Hormuz theatre.
30 U.S.-facilitated transits cannot simply be divided by a 138-vessel historical baseline. And a like-for-like series is not interchangeable with an aggregate traffic count.
An Omani lane is not automatically a safe lane.
And $84 Brent does not mean the physical disruption has ended.
The report states 80 incidents. Its annex contains 85 unique report references. Without a published counting rule connecting the two, we will not manufacture one.
Each number is real.
The mistake begins when we ask it to measure something it does not measure.