Similar-looking numbers can describe different traffic
A vessel count for the Strait of Hormuz is meaningful only when the reader knows what was counted, over what period, in which geography, and with what collection method. During the 2026 Iran war, those details became commercially decisive because public figures were often used to support claims that traffic had recovered or collapsed.
JMIC’s November 12 update illustrates the problem well. It cited a 2025 historical average of about 138 vessels per day, reported 135 US-facilitated transits over the four days from November 8 through 11, and separately discussed non-facilitated cargo traffic above 1,000 deadweight tons. JMIC expressly said the measures came from different sources and definitions. The facilitated number covered movements recorded by US Naval Cooperation and Guidance for Shipping. The other dataset used S&P Maritime and SeaVision and excluded some traffic, including vessels calling Iranian ports adjacent to the strait.
Dividing 135 by four produces an average, but it does not produce the total daily traffic through Hormuz. Adding another series without removing overlap could double-count ships. Comparing the result directly with a 2025 all-vessel baseline could create an apples-to-oranges recovery percentage.
The correct commercial question is not “What is the number?” It is “What population does this number represent, and is it comparable with the number used in our decision?”
Define the unit, period, and geographic gate
Every traffic statement should specify at least five fields: vessel population, direction, geographic gate, observation period, and source method.
Vessel population may include all detected ships, only merchant ships, only cargo-carrying vessels, ships above a size threshold, tankers, or ships participating in a coordination mechanism. Direction matters because inbound and outbound flows can diverge. Geography matters because a count at the narrow strait is not the same as a count of arrivals at Gulf ports, stationary vessels near Fujairah, or ships crossing a larger monitoring polygon.
Time also changes the answer. A four-day total can be distorted by a temporary pause or burst of movements. A monthly average smooths those events but may be too old for a voyage decision. IMO publishes average daily transit information for March through September 2026 based on several sources and labels it for general awareness. The organization also states that it no longer compiles daily transit data. Users should not extend the last monthly point to November without newer evidence.
Energy-flow measures answer another question again. Barrels per day through Hormuz describe cargo volume, not hull count. Ten laden large tankers and ten smaller or partly laden ships do not move the same volume. LNG expressed in billion cubic feet per day cannot be added directly to oil expressed in barrels. Vessel count, deadweight capacity, actual cargo, and energy-equivalent flow each support different analyses.
AIS visibility is evidence, not omniscience
Commercial vessel-tracking platforms are valuable, but their displays do not reproduce reality perfectly. AIS reception has gaps, ships may be outside coverage, static or voyage fields may be wrong, and filtering choices can omit or misclassify vessels. Duplicate identities and delayed messages can also distort automated counts.
The 2026 Gulf operating environment adds deliberate and incidental electronic interference. MARAD’s active advisory says significant GNSS interference, spoofing, and jamming continue in the Persian Gulf, Strait of Hormuz, and Gulf of Oman. It advises mariners to rely more heavily on radar ranges, visual bearings, and cross-checks against secondary systems because positional integrity is degraded.
The US Coast Guard Navigation Center’s public reports include 2026 user submissions from Fujairah, Dubai, the Gulf of Oman, and Arabian Sea describing loss or displacement of GPS positions. These are reported anomalies, not a complete census and not automatic proof of who caused them. They nevertheless reinforce the need to treat mapped positions carefully.
AIS itself is an open radio system. MARAD has warned that signals may be spoofed, producing incorrect or missing data. An analyst should therefore avoid concluding that a ship physically entered Iranian waters solely because a track jumped there, or that no ships moved because a receiver showed no tracks.
The opposite error is equally serious. Electronic interference does not make all tracking data useless. It means confidence should be graded, anomalies cross-checked, and methodology disclosed.
Build a comparison table before using a percentage
A disciplined analyst can create a compact data dictionary for each series. The fields should include source, publication time, observation window, vessel threshold, included ship types, direction, geographic boundary, AIS-dark treatment, facilitated-traffic treatment, Iranian-port treatment, and revision status.
Only series with compatible definitions should be used for direct percentage comparisons. If definitions differ, the analysis can still be useful, but the difference must be explained. For example, a count of 34 facilitated transits per day can show the scale of that channel. It cannot by itself prove total traffic was 75 percent below a 138-vessel all-traffic baseline.
Uncertainty ranges are often better than a precise percentage. If several credible sources place activity within a band but use different coverage, an analyst can describe the direction and approximate scale while withholding a false exact figure. The underlying counts and definitions should remain visible so decision-makers can judge the evidence.
Revision control matters too. Incident and traffic reports can be time-late. A daily dashboard captured at noon may be revised after a flag State or authority confirms a movement. Commercial models should store the data cutoff and refresh rules, rather than silently overwriting the value that informed an earlier decision.
Match the metric to the commercial decision
A charterer estimating available tonnage may care about ships completing outbound passage and becoming open. A cargo trader may care about loaded barrels reaching the Gulf of Oman. A terminal may care about inbound arrivals and anchorage density. An insurer may focus on exposure hours inside a defined high-risk area. None is fully answered by a generic daily vessel count.
For voyage planning, traffic density and stationary clusters affect collision exposure and bridge workload. For macro supply analysis, cargo volume and grade matter more. For a fixture, the relevant evidence may be comparable ships that received approvals and completed the route within a stated period.
Brokers should attach definitions when circulating traffic intelligence. “Thirty-four ships per day” is incomplete. “Average US-facilitated transits recorded over November 8–11, based on JMIC’s stated NCAGS series” is a usable description. It prevents a client from mistaking one operational channel for the entire market.
Zeaclub’s direction in AI shipbroking can help by storing metric definitions alongside values, rejecting invalid additions, and flagging comparisons whose populations or dates do not match. Automation should not convert a partial AIS feed into a claim of complete situational awareness. Its value is to preserve provenance and expose uncertainty so commercial teams can make faster decisions without losing analytical discipline.
This article distinguishes verified source material from ZeaClub analysis, recommendations, forecasts and hypothetical scenarios.
Sources
- JMIC Advisory Update 113, November 12, 2026Accessed 2026-11-18
- IMO Hormuz transit data for March through September 2026Accessed 2026-11-18
- US MARAD Advisory 2026-011Accessed 2026-11-18
- US Coast Guard GPS User Issue Detection and Evaluation toolAccessed 2026-11-18