Iran, Hormuz, and War as a Network

Take three news items that, read in sequence, seem to belong to different columns: a ship hit in the Strait of Hormuz, an oil license revoked by OFAC, and a technical IAEA update on access to Iranian sites. The first looks military, the second legal, the third nuclear. But when you place them on a map of dependencies, the three points begin to move together. The ship changes the price of risk; the license changes the value of a barrel; nuclear verification changes the credibility of the ceasefire that is supposed to make the corridor navigable.

This is the mental move that makes the crisis legible: do not look first at the news item; look at the channel it activates. A launch toward Israel remains, in the first approximation, a military event. A strike on Hormuz is also energy, insurance, maritime law, Gulf diplomacy, and imported inflation. A ceasefire can reduce the firing while leaving the corridor’s material problem untouched. A technical nuclear meeting can seem far from the sea until it becomes the implicit condition for reopening the sea itself.

War here does not run along a single line between states: it works as a network of chokepoints, rear areas, and negotiating channels. Iran, the United States, and Israel remain the main actors, but the real cost of the conflict is transmitted by nodes that are not all on the front: Hormuz, Qatar, Oman, Kuwait, Bahrain, the UAE, Saudi Arabia, Lebanon, the IAEA, the IMO, the UN, insurers, and energy markets. Those nodes decide when a local event becomes a regime change for the network as a whole.

The criterion is observable. A political statement counts for less than a ship that changes course, a crew that refuses transit, a war-risk cover that breaks, an IAEA inspection that cannot be verified, or an oil license that turns a barrel into a legal risk. The crisis becomes technical when these signals stop arriving separately.

From a narrative graph to a falsifiable model

The map of actors remains useful for orientation, but it cannot perform three jobs at once: describe geopolitics, identify causes, and produce a risk number. States, institutions, prices, and infrastructure are not homogeneous variables; adding their activations behind a 0–100 scale turns the analyst’s choices into a precision the data do not possess.

The model therefore begins with a stricter chain. Events alter operational risk; that risk conditions Hormuz’s effective capacity; capacity determines missing flows and rerouting; only then do prices, the energy import bill, and macroeconomic dynamics enter. An arrow is admitted only if it can be tied to data, an estimable equation, and a forecast error.

Causal chain From an event to its economic effect

The model separates observations, latent states, physical constraints, and economic consequences. Each transition needs its own data and estimate.

01 Observed
Verified events
Source provenance, attacks, blockades, mines, agreements, and measurable event features.
02 Latent
Operational risk
Escalation probabilities and transitions among normal, coercive, selective, or dysfunctional transit.
03 Estimated
Hormuz capacity
Executable traffic relative to a normal counterfactual, estimated separately for oil and LNG.
04 Physical
Missing flows
Unserved demand, bypass capacity, activation delays, and rerouting costs.
05 Market
Energy prices
Revision in expected scarcity; Brent, TTF, and insurance premia remain downstream variables.
06 Economy
Impact on Italy
Import bill, sectoral pass-through, inflation, and output with uncertainty intervals.

Define the forecast first

For each fuel $f\in\{\mathrm{oil},\mathrm{LNG}\}$ , $Q_{f,t}$ is the observed flow on day $t$ . The term $Q^0_{f,t}$ is the traffic a model trained on ordinary periods would have forecast in the absence of the crisis. It cannot be a recent moving average: once the corridor is disrupted, that average already contains the shock it is meant to measure.

The ratio between the two flows defines effective operational capacity. It does not describe the strait’s nautical width; it measures how much traffic remains executable after crews, war-risk coverage, queues, stocks, and restrictions are considered. Event $A_f(h,\delta,d;t)$ means that, within a horizon of $h$ days, capacity falls below $1-\delta$ for at least $d$ consecutive days. The main outputs are therefore an explicit probability and a loss in physical units.

\begin{aligned} Q^0_{f,t} &= \mathbb E\!\left[Q_{f,t}\mid\text{ordinary conditions}\right],\\ C_{f,t} &= \frac{Q_{f,t}}{Q^0_{f,t}},\\ p_f(h,\delta,d;t) &= \Pr\!\left[A_f(h,\delta,d;t)\mid\mathcal F_t\right],\\ L^f_{h,t} &= \mathbb E_t\!\left[ \sum_{\tau=1}^{h}Q^0_{f,t+\tau} \left[1-C_{f,t+\tau}\right]_+ \right]. \end{aligned}

Here $[x]_+=\max(x,0)$ and $\mathcal F_t$ contains only information available by $t$ . Thresholds must remain visible: a predictive publication should show several combinations of horizon, drop, and duration. For oil, $L^f_{h,t}$ is expressed in barrels; for LNG, in a consistent physical unit, without pretending that alternative oil pipelines can replace an LNG carrier.

From sources to the operational state

A war report is not automatically a true event. The database must preserve the original source, distinguish republications, and assign each episode a confirmation probability $r_i$ . Satellite imagery, maritime authorities, and shipowners can corroborate one another; ten articles derived from the same statement remain a single information chain. Observable features — distance from the route, target, tonnage, duration, damage, mines, ports, or bases — then enter the model weighted by event credibility, not by a score chosen at a desk.

Verified events feed a parsimonious daily process, for example a negative binomial model, divided into a few classes: threats to traffic, coastal or energy attacks, strikes on Gulf host nodes, coercive measures, and verified signals of de-escalation. A marked Hawkes process would make sense only with many precisely dated episodes; on a single crisis it would create more parameters than information.

Capacity is governed by a latent state $S_t\in\{0,1,2,3\}$ : normal, open but coercive, selective interdiction, generalized disruption. Transition probabilities depend on events and the previous regime; a ship strike does not add points, but it may increase the probability of moving from one state to another. The continuous value $C_{f,t}$ is inferred jointly from transit counts, AIS tonnage, waiting times, diversions, and cancellations. Satellite data from IMF PortWatch provide a useful sensor, not an infallible measure: reception, declared draft, coverage, and revisions must remain inside the model’s uncertainty.

The relevant network is physical

Downstream of the operational state, network theory takes on a narrower meaning. Oil and LNG require separate graphs with production areas, terminals, pipelines, routes, ports, storage, and final markets. Each edge has nominal capacity $\bar q_{e,f,t}$ , operational fraction $R_{e,f,t}$ , and unit cost $c_{e,f,t}$ ; on the Hormuz edge, $R_{\mathrm H,f,t}=C_{f,t}$ .

The system solves a minimum-cost flow while leaving unmet demand $u_{v,f,t}$ explicit. A bypass is therefore not counted merely because it exists on the map: it needs spare capacity, a compatible origin, available terminals, and enough time to become operational.

\begin{aligned} J^\star_{f,t} &= \min_{\{q_e,u_v\}} \left[ \sum_{e\in E_f}c_{e,f,t}q_{e,f,t} + \sum_{v\in V_f}V_{v,f}u_{v,f,t} \right],\\ s_{v,f,t}+\sum_{e\in\delta^-(v)}q_{e,f,t} &= d_{v,f,t}-u_{v,f,t} +\sum_{e\in\delta^+(v)}q_{e,f,t},\\ 0\le q_{e,f,t} &\le \bar q_{e,f,t}R_{e,f,t}, \qquad U_{f,t}=\sum_v u_{v,f,t}. \end{aligned}

$U_{f,t}$ is the physical loss the next stage must explain. The shadow price $\partial J^\star/\partial\bar q_{\mathrm H}$ measures the marginal value of additional Hormuz capacity. Min-cut also returns to its technical meaning: the sum of operational capacities crossing a partition of the physical network, not a metaphor for the number of domains involved. The EIA estimate of about 20 million barrels per day passing in 2024 and 2.6 million barrels per day of unused capacity in the main bypasses gives the order of magnitude, not an automatic substitute.

The US-Iraq agreements of July 17 matter precisely because they do not solve the short-term constraint. The deals with Baghdad, including the planned oil route through Basra, Haditha, Ceyhan, and Baniyas, indicate designed alternative capacity rather than operational capacity already available. Even if the Iraq-Syria pipeline were built for roughly 2 million barrels per day and several regional corridors reached a much larger scale by 2028, the model has to treat them as future edges with their own timelines, political risk, and bottlenecks.

Prices stay downstream

Markets react mainly to a revision in expected future scarcity. Let $M^H_{f,t}$ be the expected cumulative deficit over the next $H$ days: $\xi^H_{f,t}=M^H_{f,t}-M^H_{f,t-1}$ is the day’s new economic information. An attack that does not alter the distribution of flows may generate substantial news and little price novelty; an abrupt move toward selective interdiction can do the opposite.

\begin{aligned} M^H_{f,t} &= \mathbb E_t\!\left[\sum_{\tau=1}^{H}U_{f,t+\tau}\right],\\ \xi^H_{f,t} &= M^H_{f,t}-M^H_{f,t-1},\\ \log P_{f,t+h}-\log P_{f,t-1} &= \alpha_h+\beta_h\xi^H_{f,t} +\boldsymbol{\gamma}_h^\top\mathbf Z_t +\varepsilon_{t,h}. \end{aligned}

Jordà’s local projections estimate responses at several horizons without imposing the full dynamics of a VAR. $\mathbf Z_t$ must separate global demand, OPEC+ production, inventories, the dollar, and financial risk. Banca d’Italia research shows why identifying supply, demand, and expectation shocks is decisive. Brent, TTF, volatility, and insurance premiums therefore remain variables to explain or auxiliary sensors: they cannot sit upstream inside an index that is then used to forecast those same prices.

Transmission to the Italian economy

Italy may absorb the shock even if no ship bound for an Italian port is stopped: a change in the marginal energy price is enough. The first channel is the import bill. If $e_t=\mathrm{USD}/\mathrm{EUR}$ , a weaker euro amplifies the price in euros; freight, coverage, and rerouting enter separately through $\Delta F_t^{\mathrm{IT}}$ .

\begin{aligned} \Delta B_t^{\mathrm{IT}} &\simeq \sum_f V_{f,t}^{\mathrm{IT}} \left( \Delta\log P^{\$}_{f,t}-\Delta\log e_t \right) +\Delta F_t^{\mathrm{IT}},\\ \Delta\mathbf p &= \left(I-RA^\top\right)^{-1} RB^\top\Delta\mathbf p_E. \end{aligned}

The second line propagates energy prices through the input-output matrix: $A$ contains inter-industry inputs, $B$ exposure to imported energy, and $R$ the pass-through absorbed by prices and margins. Inflation, output, consumption, and wages then require a separate dynamic estimated through local projections and specifications that depend on the state of the economy. This is not a minor detail: Eurostat attributes 36% of Italy’s available energy in 2024 to gas, while Banca d’Italia finds energy-shock effects that vary with shock size and the inflation regime.

Uncertainty is an output

Uncertain events, unobserved states, and persistent regimes make a Bayesian estimate a natural choice. Each draw from the posterior simulates events, state, capacity, physical network, prices, and the Italian impact; the output is a probability distribution, not a privileged path. Shrinkage priors should turn off edges unsupported by data, while pooling across crises avoids learning an entire dynamic from a few months.

The model enters production only if it beats simple out-of-sample benchmarks. Disruption probabilities should be assessed with Brier and log scores; transits, flows, and prices with MAE, RMSE, and CRPS. The primary test is leave-one-crisis-out, accompanied by placebos, ablations, prior sensitivity, and temporal-leakage checks. If the full system cannot beat a seasonal transit average or a parsimonious logistic regression, the correct conclusion is to reduce it, not defend its complexity.

The map of the crisis

Separated from the predictive model, the network returns to the job it does well: showing which actors and functions can transmit a shock. The chronology is no longer a line of dates, but the movement of an event across routes, bases, insurance, negotiations, and prices. This is a descriptive structure; by itself, it produces neither coefficients nor probabilities.

From March to June, the recurring pattern is this: every military episode also shifts a civilian, energy, or diplomatic function. Iranian coercion does not have to win a conventional battle to raise the cost of the war; US and Israeli forces do not have to occupy the Gulf to make it part of the operation; regional brokers do not stop mediating when they are hit, but they mediate with less margin and more political cost.

What matters is the number of nodes a shock manages to involve in the same cycle: naval routes, bases, ports, insurance, LNG, nuclear verification, indirect channels, and the oil price. When these planes move together, the crisis is no longer only military. It becomes a network of dependencies in which every actor tries to hit the other’s systemic cost without openly taking on the cost of total war.

The observations available at 04:26 on August 23, Rome time, show an administered route war, not normalization. The central signal remains the move from claim to physical enforcement: the account of the Vela Nova cargo ship places the strike in the Gulf of Oman against a ship headed toward an Iranian port after ignored warnings. The August 14 AP roundup adds two ADNOC vessels hit by drones while crossing Hormuz on Thursday evening, with no injuries, the UAE attributing the attack to Iran, and no immediate Iranian response. On August 15 the UAE/WAM statement attributes a new attack against an ADNOC-affiliated vessel transiting Hormuz to Tehran, again with no injuries; the same-day AP roundup places it as the third attack against ADNOC-operated vessels in the past week and also records a bulk carrier hit in the hull by an unknown projectile according to UKMTO. In the August 23 check, the IMO Middle East incident ledger is updated to August 21 and remains at 68 confirmed cases and 19 seafarer fatalities, including AL WATAN, damaged in the Strait of Hormuz on August 14 with no pollution, MINOAN DIGNITY, damaged in the Strait of Hormuz on August 17 with no pollution and one fatality, and RIYAN STAR, damaged in the Strait of Hormuz on August 15 with no pollution and one fatality. The August 18 AP roundup also places a projectile strike against a vessel leaving the strait, off Oman, with engine-room damage and a casualty; the AP account of Oman and Iran adds two missiles that the UAE says were launched from Iran and fell into the sea without damage or injuries, an attribution Tehran disputes. The UKMTO recent incidents page remains at 0 current reports in the August 23 check. The UKMTO VRA overview for the week ending August 21 nevertheless keeps risk SEVERE in the Strait of Hormuz, SUBSTANTIAL in the Gulf of Oman, Gulf of Aden, and Bab el-Mandeb, IMO traffic separation arrangements suspended, and 23 UKMTO-reported projectile strike incidents since July 6. The corridor is not neutral while destination, authorization, flag, cargo, and compliance remain operating conditions.

On August 13 the legitimacy dispute becomes more explicit. The Iranian claim of control rejects the U.S. claim and ties transit to Tehran’s permission; the deadlock in U.S.-Iran talks keeps Oman and Qatar in the mediator role but does not turn a route map into ordinary traffic. Al Jazeera/Reuters report eight vessels transiting Hormuz on Tuesday according to Kpler, below the 10-day average of about 12 and far from 130-140 daily ships before the war. On the same day, the Houthi attack on Tihamah in Bab el-Mandeb kills four crew members and two rescuers or Yemeni government-allied fighters. The technical point is not to add separate theatres together; it is to observe that the Saudi alternative route and the Hormuz corridor are beginning to share the same constraint on crews, coverage, stocks, and physical protection.

On August 14 residual capacity thins further. Reuters via Al-Monitor reports traffic near standstill: nine crossings on Thursday, five on Wednesday, two Kpler-visible crossings on Friday, a separate empty LPG tanker entering the Gulf, and no oil cargo observed in the verified window. Against the pre-war baseline of more than 130 vessels per day, the distinction remains between a physically passable corridor and an economically ordinary one. The same update places the blockade tally at 62 redirected commercial vessels, three disabled vessels, and two boarded vessels; the August 14 CENTCOM update is the direct source for that operating count.

The distance between framework and normalization grows because a temporary lane is not enough to close the political regime that must make it credible. The August 10 liveblog had already crystallized the blockage: Trump described the channel with Tehran as only a partial negotiation and waited for economic pressure to rise, while Araghchi denied direct talks with Washington and tied Hormuz to the conditions already conveyed, including an end to the naval blockade. On August 11 the compensation dispute turned the route into reciprocal political accounting; the Naqvi visit to Tehran keeps Pakistani mediation alive, but does not change the threshold: ending the military campaign, the naval blockade, frozen assets, sanctions, compensation, and a regional ceasefire remain in the same cost function. On August 12 the interdiction of Vela Nova shows that this accounting produces physical effects before it produces a stabilized maritime procedure.

Between August 15 and 18 the negotiation hardens on the same political edge. The United States describes the blockade of Iranian ports as sustainable indefinitely and announces new economic pressure; Tehran continues to tie normalization to the removal of the blockade, sanctions, and frozen assets. The August 15 passage is not a reopening: Tehran has not decided to return to talks with Washington, distinguishes messages via Qatar and Pakistan from actual negotiations, and works with Oman on a temporary technical route. The Al Jazeera liveblog adds Gharibabadi’s point: Hormuz remains under Iranian control until Washington accepts Tehran’s political conditions. The 60-day MoU term expires on August 17 without an extension; the next day, AP records no scheduled talks with Tehran, a blockade still in force, and limited traffic despite the U.S. statement that the strait is open. The Iran-Oman draft therefore becomes a contested procedure: Washington rejects joint management of the exit route, threatens Oman, and does not turn the technical lane into an ordinary regime. The Iranian parliamentary committee’s approval of a plan against assets and equipment from hostile countries adds a legal and ownership risk layer on top of the route’s physical risk.

Between August 20 and 23 the political edge moves further toward economic leverage. AP records Iran’s response to U.S. economic threats: Tehran rejects the idea that sanctions and the blockade can force concessions on Hormuz, while Islamabad keeps carrying messages without reopening the expired MoU channel. Al Jazeera then places Washington’s pressure in the frame of the toughest sanctions promised by the administration, with Beijing opposing secondary sanctions against buyers of Iranian oil. The route therefore remains tied to a double cost function: physical transit risk and legal-financial risk on the cargo.

August 22 does not reopen the channel. Al Jazeera records Pezeshkian calling for an end to the war from a position of strength and an Oman-Iran call on suitable conditions for renewed dialogue and negotiations, including the implications for navigation in Hormuz; in the same frame, Washington prepares a new economic squeeze while Tehran calls it coercive pressure. The Al Jazeera liveblog then reports Trump’s claim that Hormuz is American territory. The measurable point sits elsewhere: according to the CENTCOM tally carried by the Guardian, by August 21 U.S. forces had redirected 68 commercial vessels, disabled three vessels, and boarded two. The political phrase signals claimed ownership; the tally measures selective enforcement, not normalization.

The economic edge on August 22 adds exceptions and bypasses to the same structure. AP records Mohsen Rezaei, the new head of the Supreme National Security Council, threatening neighbors that join the U.S. economic campaign and shifting possible retaliation toward oil-shipping routes outside Hormuz. In the same frame, Egypt and Iran discuss reviving the U.S.-Iran channel; Iraq and Iran say some Iraqi oil tankers have been facilitated, without a public number of vessels or a clarified quid pro quo; France and Saudi Arabia prepare discussions on Omani ports outside the Persian Gulf, pipelines, and rail links. These are three different signals of the same mechanism: the strait does not stop mattering when some ships pass, because operating value moves into permits, sponsors, routing, and bypass capacity.

The material base confirms the caution: physical interdiction of Belma, a blockade tally updated on August 21 with 68 redirected commercial vessels, three disabled vessels, and two boarded vessels, the new ADNOC-affiliated vessel identified by the UAE on August 15, the bulk carrier hit in the hull according to UKMTO, KAVOMALEAS, ACHELOOS, KAIFAN, GASLOG SHANGHAI, MINOAN PIONEER, VELA NOVA, AL WATAN, MINOAN DIGNITY, and RIYAN STAR in the IMO ledger, 23 UKMTO-reported projectile strike incidents since July 6 in the VRA overview for the week ending August 21, the July 21 attack with crew abandonment, Houthi attacks on Encelia and Layla, the Tihamah strike with six deaths among crew and rescuers, pressure on Yanbu, Jizan, and Bab el-Mandeb, at least 6,000 seafarers stranded on about 400 vessels around the strait, U.S.-Saudi strikes in Iraq, the July 29 CENTCOM wave against IRGC targets in Iran, five Iranian missiles intercepted by Jordan, drones over Kuwait, and no new public IAEA access. AP/Lloyd’s List put traffic at 84 transits during the week of July 27-August 2, up from 45 the previous week but still far below more than 700 in a typical pre-crisis week; the next MarineTraffic signal worsens the daily reading, with crossings falling from 15 on Friday to 11 on Saturday and 6 on Sunday, 17 vessels on the Iran-designated route, and 10 on a route classified as undetermined. The Kpler reading reported on August 12 shifts the comparison again: eight Tuesday transits, a 10-day average of about 12, and a pre-war baseline of 130-140 daily crossings; the August 14 Kpler/Reuters reading brings it to traffic near standstill, with two visible Friday crossings and no observable oil cargo. The new August 18 AP/Kpler reading measures 95 crossings in the previous week, down 19.5%, only three Sunday crossings, visible traffic on the Iran-designated route, and no crossings on Omani routes; Bab el-Mandeb rises to 254 transits, while dark transits fall from 40 to 16. The subsequent Al Jazeera/Kpler reading reclassifies August 1-19: 236 total transits, 148 dark or unclassified, 83 on the Iran-designated route, three on Omani routes, and two on the pre-war central channel; among 112 oil-and-gas vessels, 89 remain dark or unclassified, 21 use the Iranian route, and two use the Omani route. The data do not measure reopening; they measure the distribution of risk across AIS visibility, compliance, and lane choice. UKMTO Warning 108-26 remains the verified warning that places an August 8 projectile strike 18 nautical miles east of Khasab, with the fire extinguished, the vessel and crew safe, and no pollution reported. The August 14 UKMTO VRA overview keeps IMO traffic separation arrangements suspended, estimates flows about 90% below pre-conflict baselines, and concentrates 16 of 18 projectile strike incidents since July 6 on the southern Omani route, with 20 UKMTO projectile strike incidents over the same period and 76 year-to-date total events. The August 21 UKMTO VRA overview keeps the same traffic separation procedures suspended, raises the count to 23 UKMTO-reported projectile strike incidents since July 6, and keeps risk SEVERE in Hormuz. The change is incremental but material: more events in the route pressure register, not a populated recent-incidents page or a new IMO count. IMO, updated to August 21 and checked on August 23, remains at 68 confirmed incidents and 19 seafarer fatalities, including VELA NOVA 70 nautical miles south of Gwadar, Pakistan, damaged with no pollution, AL WATAN in the Strait of Hormuz on August 14, damaged with no pollution, MINOAN DIGNITY in the Strait of Hormuz on August 17, with one fatality and no pollution, and RIYAN STAR in the Strait of Hormuz on August 15, with one fatality and no pollution. The UKMTO recent incidents page checked on August 23 remains at 0 current reports. The warning, UKMTO overview, recent incidents page, and IMO ledger measure different risks: safety alert, route pressure, current publication, and confirmed incidents.

August 13 adds an environmental and logistical cost to the route constraint. AP records cleanup work for a spill off Qeshm, with no independently determined cause, and a second Omani spill tied to Caroline Bezengi, a sanctioned vessel grounded after a June explosion. The same roundup records Houthi-government clashes in Taiz, a new drone strike against Jizan, and a U.S. claim that the blockade can be sustained; the George Washington arriving in the Arabian Sea, with Lincoln leaving after more than 240 days of continuous deployment, shows that the blockade requires rotations and naval capital, not a simple legal formula. The new Task Force Falcon Strike also moves the military layer toward one-way drones above, on, and below the surface. That is an enforcement and deterrence capability, not a reopening signal. The August 15 AP roundup adds the Houthi attack on Mokha port, with three deaths and suspended operations: the Hormuz constraint remains connected to the ability of Red Sea alternative routes to absorb or amplify the shock.

The new ADNOC episode reported by the UAE moves that cost onto a national operator at the most sensitive point in the network. After the two August 14 attacks, AP describes the August 15 case as a third attack against ADNOC-operated vessels in the past week; Anadolu reports the UKMTO detail of the bulk carrier hit in the hull by an unknown projectile, with the crew safe and damage not yet quantified. The events produce no casualties and do not by themselves identify a permanent change in the transit regime, but they raise the implicit insurance premium for cargoes, crews, and sovereign sponsors already tied to the corridor.

The exit of the QatarEnergy-controlled LNG tanker Al Areesh remains the first visible transit of that type since July 11. The move from 45 to 84 weekly transits measures residual capacity under administration, not ordinary reopening; the daily decline recorded over the weekend shows how reversible that capacity remains. Brent moved above USD 100 at the July 23 settlement, fell in the July 28 intraday continuous quote, rebounded to USD 91.04 at 22:59 BST on July 31, fell to USD 80.07 at 14:25 BST on August 5, settled at USD 83.55 on August 7, rose to USD 87.72 on August 11, and MarketWatch marks it at USD 88.55 at 22:59 BST on August 12, with an August 12 settlement at USD 88.98, then at USD 86.93 at 21:20 BST on August 13, USD 88.82 at 22:58 BST on August 14, with an August 14 settlement at USD 88.52, and USD 91.53 at 9:53 BST on August 19, with an August 18 settlement at USD 91.02, then USD 93.60 at 5:32 BST on August 21, with an August 20 settlement at USD 93.78, and USD 92.22 at 22:59 BST on August 21, with an August 21 settlement at USD 92.67; WTI continuous moves from the USD 78.18 August 7 settlement to USD 82.77 at 18:15 EDT on August 12, with an August 12 settlement at USD 83.27, then to USD 81.21 at 16:59 EDT on August 13, USD 82.40 at 16:59 EDT on August 14, with an August 14 settlement at USD 82.40, and USD 84.56 at 4:17 EDT on August 19, with an August 18 settlement at USD 84.06, then USD 86.20 at 1:06 EDT on August 21, with an August 20 settlement at USD 86.83, and USD 86.64 at 16:59 EDT on August 21, with an August 21 settlement at USD 87.06. The rise recorded by Al Jazeera already read the price as a risk premium on Iranian conditionality; the August 11 EIA STEO moves the frame to estimated scarcity: Hormuz at 4.9 million barrels per day in 2Q26 versus 21.6 in 4Q25, average shut-ins of 5.5 million barrels per day in July, flows severely constrained through August, and 3Q26 Brent around USD 85/b, USD 11 above the previous estimate. The EIA WPSR is updated to the August 19 release for the week ending August 14 and adds a stocks, refining, import-export, and apparent-demand confounder in the same price window. The August IEA Oil Market Report moves the same picture to global scale: 2026 demand cut to -1.6 million barrels per day, July supply at 101.5 million barrels per day, and observed stocks down 410 million barrels since the war began. These measured facts are useful signals. They are not a calibrated probability of further disruption or an estimate of Italy’s loss.

Descriptive map Actors and dependencies in the crisis

Coercion
US-Israel alliance
Mediation
Hosting and rear area
Energy
Verification / rules
Regional spillover

This map orients the reporting; it does not estimate probabilities. The United States-Israel edge binds alliance and military capability; Kharg locates Iranian exports; Hormuz concentrates the chokepoint; Qatar and Oman show diplomatic channels; the Gulf nodes form the exposed rear area. The predictive model uses a separate physical network with capacities, costs, and oil or LNG flows.

The United States-Israel edge

There is one line in the graph that should not be read like the other fire edges. The United States and Israel are connected by targeting, intelligence, ammunition, missile defense, diplomatic cover, industry, and regional hosting. Israel’s conventional advantage depends on American depth; American freedom of action depends on the resilience of Israel’s rear area and of its Gulf partners.

The relationship remains asymmetric. Israel can maintain tempo and pressure on the Iranian target set because Washington provides capability, industrial scale, and a regional guarantee; Washington pays for every escalation in terms of exposed bases, spent ammunition, allies to reassure, and the political cost of maritime protection. In the graph, that edge is the hinge that turns a military campaign into a coalition problem.

Why Hormuz dominates the picture

Hormuz dominates because it is small. In a network, a narrow point does not have to be large to be powerful: it has to sit on the path of many functions. If that point becomes uncertain, the system does not merely lose a route; it loses a simple way to coordinate prices, ships, stocks, insurance, and military guarantees.

Four balances overlap here. The first is military: the corridor can be mined, escorted, blocked, or reopened for designated routes. The second is energy: a critical share of oil and LNG passes through the Gulf. The third is insurance: freedom of navigation exists only if shipowners, crews, and war-risk coverage accept the transit. The fourth is diplomatic: no actor wants to own the political cost of a permanent naval mission alone.

The EIA estimates for Hormuz set the scale: in 2024, roughly 20 million barrels per day of oil and refined products and about one-fifth of global LNG trade passed through the corridor; unused capacity on the main Saudi and Emirati bypasses was around 2.6 million barrels per day. Stocks can buffer oil faster than they can buffer shipping: the IEA has indicated an emergency response of roughly 426 million barrels, but that buffer cannot replace LNG, crews, war-risk coverage, or the physical protection of the corridor.

By June, the word open was already insufficient. A corridor can reopen and remain coercive. It is enough to narrow routes, impose vetting, raise the insurance premium, make crews uncertain, or force every ship to ask what guarantee it is actually buying. Brent can fall after a ceasefire because it prices a probability; ships, by contrast, have to cross a field of rules, mines, coverage, and military credibility.

Between June 26 and July 17, the corridor returns to a logic of enforcement. CENTCOM records attacks on Ever Lovely and Kiku and responds with strikes on missile depots, drones, and coastal radar; AP places new separate talks in Qatar; the dispute over the IRGC route becomes propaganda and operational control; attacks on Al Rekayyat, Wedyan, and Cyprus Prosperity bring three commercial ships back to the center of the picture; OFAC revokes the oil license and limits the wind-down to July 17; US strikes on July 8 hit additional coastal targets; on July 10, Washington asks Tehran to publicly declare Hormuz open and free of attacks, while Tehran claims exclusive control over the strait; on July 11, CENTCOM records a third round of roughly 140 targets and more than 300 targets hit during the week; on July 12, the hit Cypriot container ship, the evacuated crew, the missing seafarer, and Iran’s declaration of closure move the sea from a negotiating condition back to an operational theater. The next jump comes with the new CENTCOM round on July 13, which hits Bushehr, Chah Bahar, Jask, Konarak, Abu Musa, and Bandar Abbas, and with the blockade of traffic to and from Iranian ports reactivated on July 14 at 16:00 ET. The ceasefire breaks across shipping, strikes, licenses, route control, Iranian ports, and host nodes: the corridor once again measures real trust between the parties faster than any communique. July 15 adds the shift from declared blockade to material enforcement: AP reports the tanker Belma disabled by Hellfire missiles while heading toward Kharg, two commercial vessels diverted after US contact, a strike on Greater Tunb, and additional strike waves the same day. Hormuz is no longer only the point where the parties argue over who may authorize passage; it is the point where authorization is enforced against hull, route, insurance, and port counterparty.

July 16 does not change the crisis structure, but widens its surface. AP reports US strikes as far as Tehran, Semnan, Qeshm, and Bandar Abbas, alongside Iranian retaliation against Bahrain, Jordan, and Kuwait; the same day leaves the latest CENTCOM operational confirmations anchored to July 15, when Belma is disabled and a new strike wave hits command, air defense, missile, drone, and coastal surveillance capabilities. The operational signal remains the same: the route is not closed indiscriminately, but every passage tied to Iranian ports becomes a choice of compliance, insurance, and force.

The next step makes the corridor less maritime in the narrow sense and more infrastructural. The CENTCOM release of July 16 places the sixth consecutive night of strikes on coastal surveillance, air defense, military logistics, and maritime capabilities; AP then reports Saturday’s exchange of strikes and attributes to CENTCOM a seventh consecutive night, with targets on surveillance, underground storage, logistics, and maritime capabilities. Bridges in Hormozgan, links toward Bandar Abbas, and the Chabahar tower move the cost from transit alone to the space that makes transit administrable. The tanker hit on the route closest to Oman, with minor damage and no injuries according to UKMTO as reported by AP, keeps the central signal: the sea remains usable only as long as the network serving it accepts risk, timing, and rules of engagement.

July 18-25 shifts the center of gravity from enforcement against individual transits to the resilience of the campaign. CENTCOM confirms the eighth consecutive night of strikes, with targets on coastal surveillance, air defenses, maritime capabilities, and missile or drone storage; the next release closes the ninth night against command centers, air defenses, coastal surveillance, maritime capabilities, missile or drone sites, and communications networks. AP then reports the start of a tenth night, Hormuz traffic largely stalled, seven vessels redirected, and one disabled since the blockade returned. By July 20, the maritime picture is more verifiable: the IMO updates confirmed incidents to 59 and includes KAVOMALEAS, damaged and abandoned on July 19 eight nautical miles northwest of Kumzar, Oman. Even a corridor that is not formally closed can become hard to use if every transit has to price attack risk, ambiguity over the source of damage, and uncertainty over who guarantees the route.

July 21 pushes that state toward more persistent dysfunction. AP reports a tanker attacked in the strait off Oman and abandoned by its crew; the IRGC also claimed two July 20 attacks against vessels in the same passage. The new IMO count brings confirmed incidents to 61 as of July 21: ACHELOOS is listed as damaged 17 nautical miles east of Dibba, UAE, while KAIFAN is listed as damaged and abandoned 8 nautical miles northeast of Limah, Oman, with two seafarers injured. The technical difference matters: a formally total closure is not needed for effective capacity to fall; it is enough that even the route encouraged around Oman becomes a choice about crew, cover, and rescue probability.

The July 22 step clarifies the tension between formal openness and effective capacity. CENTCOM places the 11th consecutive evening of strikes at 8:15 p.m. ET on July 21, with targets on military operations centers, maritime capabilities, aircraft hangars, drone storage, and logistics. In the same release, the command says the strait remains open and claims roughly 900 assisted commercial transits since early May, equal to 450 million barrels of crude. The operational reading is harsher: the AP update puts redirected commercial vessels at eight and keeps one vessel disabled since the blockade returned. Legal-military openness therefore does not mean an ordinary corridor; it means transit under escort, selection, and risk pricing.

On the European night of July 23, the CENTCOM release for July 22 brings the campaign to the 12th consecutive evening, with targets on maritime capabilities, missile and drone storage, coastal surveillance sites, and air defenses. Redirected vessels rise to nine, with one still disabled to prevent entry to or exit from Iranian ports. The point is not only quantitative: each increment in enforcement makes it harder to separate formal freedom of navigation from operators’ effective ability to buy coverage, crew, and credible protection.

July 24 changes the weight of the human constraint. The War.gov/CENTCOM release places the 13th consecutive night of strikes at 9 p.m. EDT on July 23, with targets on command centers, drone storage, communications networks, coastal surveillance, and maritime capabilities. The command repeats that the strait remains open with U.S. military support; the same day, AP reports the UN/OHCHR call for help for at least 6,000 seafarers stranded on about 400 vessels around Hormuz, some without essential supplies or family communications. Effective capacity therefore does not depend only on mines, missiles, and the designated route: it depends on the concrete ability to keep crews aboard, assure rescue, and prevent a formally open transit from becoming an abandonment problem.

July 25 introduces an operational pause that does not amount to normalization. The CENTCOM public calendar remains stopped at the 13th night and AP reports no new overnight strikes reported in Iran; the same update keeps tension over the strait high because the blockade of Iranian ports, the ship fired on in the Gulf of Oman, and threats against U.S. bases in neighboring countries remain in the same window. A 24-hour pause lowers the campaign’s tempo; it does not by itself rebuild insurance cover, crew availability, or trust in the transit mechanism.

The Gulf as an exposed rear area

Now widen the ring by one step. The word rear area can mislead: in the Gulf it does not mean a safe space behind the front, but the collection of bases, ports, airports, refineries, commercial hubs, and diplomatic brokers that makes the campaign sustainable. Kuwait, Bahrain, Qatar, the UAE, and Saudi Arabia are not just geography around Iran. They are the nodes that turn a local hit into a coalition cost.

Kuwait condenses the problem well: military hosting, port infrastructure, refining, a civilian airport, and the relationship with Washington sit in the same node. Shuaiba, Mina Al-Ahmadi, and Kuwait International Airport make the same chain visible: retaliation against a host node can touch soldiers, jets, refining, civilian flights, and a diplomatic response within a few hours.

Qatar is even more delicate because it combines Al Udeid, LNG, North Dome/South Pars, mediation, and territorial vulnerability. When Doha hosts separate US-Iran talks and, in the same cycle, a Qatari tanker enters the sequence of attacks, the broker stops being an abstract channel: it becomes a node that has to defend territory, energy, diplomatic reputation, and industrial continuity.

From July 12 to 18, the rear area stops being a concept and becomes a list of interceptions, alerts, and collateral damage. In Qatar, three people remain injured by shrapnel after Iranian fire is intercepted; the UAE activates defenses against missiles and drones; Bahrain sounds sirens, Kuwait reports interceptions, and Oman records strikes in the northeast facing the strait. With the blockade reactivated, AP reports new alerts in Bahrain and Kuwait, exchanges of fire in the strait, and dozens of Iranian missiles and drones directed at Arab Gulf states. The important point is not that every Iranian claim has already been verified at the same level: it is that the US response on Hormuz pulls brokers and host nodes into the immediate retaliation calculation.

The next day extends the same logic: Jordan says it shot down three Iranian missiles, while the IRGC threatens that regional oil and gas exports will be “for everyone or for no one.” The phrase should not be read as a mechanical forecast of total closure; it matters because it turns the blockade on Iranian ports into an explicit threat against the Gulf’s energy function. On July 17, with Qatar, Kuwait, Jordan, Bahrain, and Iraq inside the retaliation perimeter reported by AP, the Gulf stops being only a logistical rear area: it becomes the political multiplier of every strike on the corridor. Missile debris wounds a child in Qatar, a Kuwaiti desalination and power facility is damaged, Jordan intercepts new missiles, Bahrain sounds sirens, and drones over Irbil pull Iraq back into the same rear-area calculation. The weekend jump adds direct military cost: CENTCOM confirms two US service members killed in Jordan on July 17, unidentified remains at the same site, and one service member killed in Iraq on July 18 during the controlled detonation of a downed Iranian drone. In the same cycle, AP reports a second day of attacks on water facilities in Kuwait, a Kuwaiti oil facility hit, intermittent airspace closure, and active defenses in Jordan, Bahrain, and Saudi Arabia. The rear area no longer absorbs only political risk: it converts the cost of the route into casualties, essential services, and operational continuity.

July 20 adds a second chokepoint to the same network. AP reports three Iranian missiles shot down by Jordan, active Kuwaiti defenses, and Bahrain’s condemnation of drones directed at air-traffic systems. In the same cycle, the Houthis announce a maritime embargo against Saudi shipping at the passage between the Red Sea and the Gulf of Aden. Bab el-Mandeb does not replace Hormuz in the model; it duplicates the network risk, because it can compress the Saudi redundancy toward the Red Sea while the Gulf corridor remains fragile.

The following day, that duplication stops being only declaratory. The U.S.-Pakistan condemnation moves Bab el-Mandeb into the crisis’s political perimeter, while Houthi-controlled SABA says six ships were redirected in the Red Sea after the group’s warnings. AP does not confirm the number, but MarineTraffic data cited in the same update show at least three tankers making U-turns. The datum requires the caution owed to a partisan source, but it still enters the observational chain: if even an unconfirmed threat produces deviations, the cost of the alternative corridor begins to move before physical damage. The State Department worldwide caution pushes the same risk onto a wider scale, because it signals vulnerability for interests and places associated with the United States outside the immediate theater as well.

On July 23, Bab el-Mandeb stops being theoretical redundancy. AP reports Houthi-claimed attacks on the Saudi tankers Encelia and Layla, with fires and no reported casualties; Saudi Press Agency reports the fire on Encelia and UKMTO records a projectile strike 130 kilometers southwest of Al Shuqaiq. The technical consequence is that Saudi redundancy toward Yanbu becomes a vulnerable edge, not a neutral bypass: if Hormuz reduces Gulf exit capacity and Bab el-Mandeb raises risk and costs in the Red Sea, the model can no longer treat the two passages as separate shocks.

July 24 shows why the rear area is not just geographic context. AP reports explosive drones over Irbil, where U.S. forces operate, sirens in Bahrain with no immediate damage reported, and seven Iranian missiles plus six drones shot down by Jordan. U.S. strikes also hit an IRGC naval base in northern Iran, Qeshm, Isfahan, Khuzestan, and Fars according to Iranian media reported by AP. In network terms, the map widens in two directions on the same day: Bab el-Mandeb remains the vulnerable bypass of the Saudi route, while Irbil, Bahrain, and Jordan measure whether the coalition can sustain the corridor without turning every host node into a permanent front.

July 25 pulls Saudi redundancy even more clearly into the network conflict. AP reports Houthi-claimed attacks on Aramco facilities in Yanbu and Jizan, repeated Saudi alerts in both cities, and Saudi strikes on Hodeida after the week’s attacks on Saudi tankers in the Red Sea. The consequence is not that Bab el-Mandeb becomes more important than Hormuz; it is that the bypass used to route around Hormuz acquires its own military, insurance, and political cost.

July 26 brings a pause, not a regime. AP reports the halt in U.S. strikes after nearly two weeks, no new Iranian attacks on U.S. bases since Friday night, and several Iran-Oman technical rounds on traffic through the strait; in the same frame, the U.S. naval blockade remains operational and AP points to a second commercial ship disabled after trying to breach it. Al Jazeera reports from Iranian media that a tanker exploded after hitting a mine in the Strait of Hormuz. The operational distinction matters: the pause slows the strike tempo, but it does not restore ordinary capacity while transit, enforcement, mines, and route rules remain subject to bargaining and coercion.

July 27 tightens that distinction. AP reports three days without new U.S. or Iranian attacks, progress by mediators led by Qatar and Pakistan, and Friday-Saturday Iran-Oman talks on the strait mechanism; in the same update, however, Tehran says Hormuz remains closed and that the situation has not changed. The U.S. blockade is still producing physical effects: 17 commercial vessels have been redirected, two disabled, and two boarded; commercial traffic through the strait has fallen to a three-week low, while no new maritime attacks have been confirmed in the past 72 hours. This is the pause’s most useful form analytically: kinetic risk recedes before effective capacity does.

July 28 does not add a new strike night to CENTCOM’s public calendar, but it tightens the measurements of effective capacity. The IMO raises confirmed incidents in the Middle East to 62 as of July 27 and puts seafarer fatalities at 17; MarketWatch records Brent continuous at USD 84.34 at 08:13 BST on July 28, with WTI continuous at USD 80.26 at 04:35 EDT. The reading is unchanged: the market can reduce the panic premium faster than crews, insurers, and transit rules can rebuild ordinary capacity.

July 29 breaks the language of pause. CENTCOM records U.S.-Saudi strikes in Iraq against logistics and weapons sites of Tehran-aligned groups, in response to more than 30 IRGC-directed drone attacks in the previous 72 hours against U.S. forces and Saudi energy infrastructure. Petra reports five missiles launched from Iran and destroyed by Jordanian air defenses; AP adds that missiles aimed at U.S. forces were intercepted. On the maritime corridor, the signal remains split: Iran’s foreign ministry had confirmed after talks with Oman that strait traffic had not changed; Al Jazeera reports Iran’s rejection of Muscat’s proposed equal management formula and the IRGC claim that three tankers were struck or stopped. As long as IMO remains at 62 confirmed incidents as of July 27, those tankers belong to the field of operational claims awaiting verification, not to the confirmed count. MarketWatch records an intraday rebound, with Brent continuous at USD 84.82 at 07:11 BST and WTI continuous at USD 82.05 at 01:36 EDT on July 29: the market corrects the pause, but it still does not isolate the physical Hormuz shock.

July 30 moves the break in the pause from the network’s rear area back onto Iranian territory. CENTCOM places a new heavy wave of U.S. strikes at 10 p.m. ET on July 29 against dozens of IRGC targets in Iran, including command centers, missile and drone facilities, coastal surveillance and defense sites, and maritime capabilities. AP reports five Iranian missiles intercepted by Jordan with no casualties, while Al Jazeera places part of the target set in southern Iran, the Gulf, and around Hormuz, and attributes injuries and people trapped on Qeshm to Iranian officials. The regional jump remains visible in the rear area: AP describes Saudi participation in Iraq strikes, possible drone attacks on two gas vessels at Damietta, and no imminent direct diplomacy. At sea the signal is subtler: Reuters/Kpler/LSEG as reported by the Guardian record the exit of the QatarEnergy-controlled LNG tanker Al Areesh, the first visible transit of that type since July 11, and 12 commodity ships passing on Wednesday. That is proof of residual capacity under risk, not the return of an ordinary corridor.

July 31 and August 1 add no confirmed ordinary reopening. AP reports that there were no overnight U.S. strikes on Iran but the White House was weighing new military options, and then reports new U.S. strike threats, attacks over Kuwait, a tanker hit, and eight Saudi ships forced by the Houthis to reroute from Bab el-Mandeb. UKMTO separately records a tanker struck by an unknown projectile 11 nautical miles northeast of Lima with engine-room damage and no casualties, while a second warning places a splash and explosion close to a tanker 21 nautical miles northeast of Khasab, without reported damage. Kuwait’s position, carried by WAM/KUNA, adds attacks against vital installations and civilian vehicles on Boubyan Island. The Lima warning then enters the IMO ledger as GASLOG SHANGHAI, damaged on July 31 with no pollution, while Khasab remains separate until it becomes a confirmed incident. Tehran’s position remains coercive rather than conciliatory: Al Jazeera records Hormuz as leverage, no direct request to Washington, and messages sent through mediators. The distinction matters: an UKMTO warning changes operational risk; the IMO count changes the confirmed incident ledger; IRGC claims change the threat environment but still require maritime verification.

August 2 shifts the center of gravity from retaliation to the convertibility of the pause. AP reports that Trump wants to hold off new strikes because regional allies had reached the parameters of a deal; the proposal would reopen the strait, halt regional attacks, end the U.S. naval blockade, and allow Iranian oil exports, but it remains an unconcluded frame. Iran’s refusal to return to the pre-February 28 status and the no breakthrough recorded by Al Jazeera keep the corridor inside a logic of authorization: the pause lowers the potential tempo of strikes, but does not by itself rebuild ordinary transits, insurance coverage, the IAEA chain, or trust in the Oman-Iran mechanism.

August 3 makes the window more measurable without closing it. AP reports that Trump says talks will begin Monday, while Tehran denies direct talks with Washington and keeps the active channel with Oman; in the same update, Kpler counts nine vessels transiting Sunday, mostly through an Iranian-controlled route. The datum moves the reading away from binary closure and toward administered capacity: passage exists, but inside a politically assigned geometry. The Al Jazeera frame on the Oman-Iran route in final stages remains consistent with the mechanism; the oil decline shown by MarketWatch and AP measures negotiating hope, not proof of ordinary transit.

August 4 turns that geometry into a more explicit negotiating draft. AP reports progress on an Iran-Oman formula with Gulf entry through an Iranian-controlled route, exit through an Oman-controlled route, service fees, and removal of the U.S. blockade on Iranian ports as the political condition. Al Jazeera places the same talks inside safe inbound and outbound lanes, while Tehran keeps denying direct talks with Washington. Oil repricing enters the same sequence: MarketWatch ties the decline to the possibility of a deal, but the new warning on a cargo ship hit off Oman prevents reading it as proof of normalization.

The UAE shows the civilian-logistical side: Jebel Ali, Khalifa, Fujairah, airspace, desalination, and commercial continuity. Bahrain converts local damage into political and legal coordination: the CENTCOM meeting with 12 nations at the start of July matters because it moves Hormuz from the level of a single incident to the level of regional management. Saudi Arabia measures redundancy: Yanbu and pipelines reduce dependence on Hormuz, but concentrate new vulnerability if they become targets or bottlenecks.

In the Gulf, war looks less like a political map and more like a circuit. A hit node matters according to the number of functions it forces actors to recalculate: defense, insurance, transit, diplomacy, prices, allies’ reputations, crew availability, and voters’ tolerance.

The nuclear issue is not a separate chapter

The nuclear issue looks like another film: verification, enrichment, reports, access, the IAEA’s technical language. In the graph, however, it is a slow layer resting on top of the sea’s fast layer. The sea decides whether pressure is immediate; the nuclear layer decides which concessions can be exchanged for which guarantees. If Tehran links the reopening of Hormuz to the end of the blockade and Washington links sanctions relief to verification and limits, sea and nuclear become the same negotiation seen from two different sides.

The IAEA page of Iran reports is useful precisely because it slows the pace of the news. It reminds us that verification is not a political communique: it requires access, continuity, a documentary chain, and a minimum of operational trust. When the IAEA becomes a central node, that does not mean it can impose an agreement. It means that without credible verification, an agreement is less convertible into real de-escalation.

The June-July sequence shows the point: initial signature, IAEA technical work, dispute over access to sites, Swiss talks, renewed pressure on Hormuz, then an IAEA Board resolution on urgent cooperation, complete information, and access. The nuclear issue cannot be separated from the sea because material inventory, site traceability, and sanctions relief enter the same political transaction. Isfahan measures the material side of the problem; Hormuz measures how credible that transaction is outside the negotiating rooms.

On July 10, the connection becomes even more explicit: the United States links a possible agreement to the delivery of the highly enriched stockpile and to an end to attacks in the strait. On the same day, the Treasury targets Ali Ansari, Smart Global Limited, and Iranian currency-exchange networks, granting Smart Global only a wind-down window through August 9. The nuclear issue no longer passes only through access and inventories: it also passes through the bankability of the counterparties that would have to support an agreement.

The Treasury package of July 15 moves the same mechanism into procurement: seven individuals and entities are designated for an international network supporting IRGC weapons purchases, and the action is tied to the non-proliferation authority of E.O. 13382. This is not a new nuclear inventory datum, but it changes the graph’s legal layer: the maritime crisis also produces counterparty risk across aviation, transport, finance, and military components.

Also on July 10, the Security Council returns the question to its most technical form: the IAEA has lost continuity of knowledge over decisive parts of Iran’s declared program, from centrifuge and component inventories to heavy water and uranium concentrate. This is not a side note to Hormuz. If the maritime crisis demands immediate passage guarantees, the nuclear layer demands a documentary chain that cannot be rebuilt with a single political sentence. The same ceasefire can therefore reduce the firing and remain fragile because it still does not know how to turn stockpiles, access, and routes into one verifiable whole.

The reactivation of the blockade makes this coupling more visible. The interim agreement left a 60-day negotiating period open; from August 17 that window has expired without a public extension, and if the sea returns to being administered through blockades, escorts, and threats of a toll, the nuclear negotiation loses its technical space and becomes part of the price of access to the corridor. As of August 22 there is no new public IAEA access that changes the constraint: the Board Reports page remains at the February and June reports, while the Resolutions page remains at GOV/2026/40 of June 10. The material fact is the discontinuity of knowledge, not a new estimate of the stockpile. Pakistani mediation, the Oman-Iran technical rounds on the strait, Trump’s announced pause, and the new negotiating frame have lost the MoU’s time buffer and can open or close a diplomatic window, but they do not by themselves rebuild inventory, design information, or the access chain. Iran’s allegation of a US strike on the under-construction Darkhovin nuclear site does move the risk layer: AP reports that the IAEA described the site as very early-stage and without nuclear material at its last visit, so the issue is not yet a verified radiological release but a further compression of access, design information, and technical trust. The collapse of the interim deal does not add a nuclear datum, but it removes the frame that was supposed to hold together an operational stop, the reopening of Hormuz, and negotiation on the program.

The new U.S. threat around Pickaxe Mountain tightens the link between verification and the route. The technical profile of the site places it south of Natanz, still under construction and never inspected by the IAEA; what Tehran has already moved under the mountain remains unclear, while there is no publicly verified restart of enrichment after the 2025 strikes. The operating sentence remains the same: follow the material. Without IAEA access and continuity of knowledge, a threat against a site does not by itself become a stockpile estimate; it does, however, add pressure to the channel that should separate maritime reopening, nuclear negotiation, and military deterrence.

Compressed timeline

The chronology should be read as a sequence of activations, not as a list of incidents. The question for each row is concrete: which event altered routes, capacity, coverage, verification, or bargaining conditions, and through which channel could that change reach the next stage?

Phase	Mechanism	Signal to remember
2025–early 2026	The nuclear and sanctions dossiers prepare the ground.	UN snapback, IAEA requests for full inspections, and Iranian warnings about regional war place Hormuz in the vocabulary before the attack.
February 28–March 6	The military trigger becomes regional immediately.	Joint US-Israeli attacks, Iranian response across the Gulf and Israel, Port Shuaiba, US advisories for Qatar and the UAE, and IMO/JMIC in crisis mode.
March 8–21	The war hits leadership, civilians, energy, and nuclear-sensitive nodes.	Iranian succession, displacement from Tehran, South Pars/North Dome, the Bushehr alert, and pressure on Qatar, Kuwait, the UAE, and Saudi Arabia.
March 22–26	Markets anticipate more than physical flows.	US tactical pause, IEA stocks, an attempt at a naval coalition without full ownership, an IRGC-vetted route, and higher insurance costs.
April 7–12	The ceasefire becomes layered.	Pakistan and Qatar enter the channel; Lebanon remains outside the perimeter; Hormuz passes only a few ships, with mines, a US mine-clearance mission, and competing narratives.
April 13–29	The corridor reopens without becoming neutral.	US blockade of Iranian ports, routes designated by Tehran, fire on commercial traffic, India involved diplomatically, the UN on maritime security, and the Isfahan node.
May 3–June 15	Reopening becomes administered and coercive.	Project Freedom, US naval escorts, 50 then 100 ships redirected, Kuwait International Airport hit, new strikes, and an initial US-Iran agreement.
June 16–25	The signature shifts to implementation.	The IAEA, Swiss talks, the Oman/IMO route, seafarer evacuation, and a cargo ship hit in the strait show that the agreement still does not control the corridor.
June 26–July 2	The route becomes an enforcement object again.	Ever Lovely and Kiku bring CENTCOM strikes; Doha hosts separate talks; the Arista ship undercuts Iran’s narrative of a compulsory route.
July 6–9	The ceasefire enters an intermittent regime.	Al Rekayyat, Wedyan, and Cyprus Prosperity, more than 80 US targets hit, a new strike wave, OFAC GL X1, Brent above USD 78, alerts in Kuwait and Bahrain, and the IAEA Board on access and stockpiles.
July 10	The crisis moves from retaliation to a public constraint.	Washington asks Tehran for public guarantees on Hormuz; Tehran claims exclusive control of the strait; OFAC targets Ali Ansari, Smart Global, and Iranian exchange houses; Brent closes at USD 76.01 but remains above the week’s start.
July 11–12	The route becomes a test of the regional rear area.	A third CENTCOM round of roughly 140 targets, more than 300 targets during the week, a Cypriot container ship hit, Hormuz declared closed by the IRGC, Qatar injured by shrapnel, and defenses and alerts in the UAE, Kuwait, and Bahrain.
July 13	The blockade returns from threat to procedure.	CENTCOM announces a blockade of traffic to and from Iranian ports from 16:00 ET on July 14; Trump proposes a 20% toll; AP records a new US round and Brent rises to USD 83.30.
July 14-28	Route war becomes physical, infrastructural, humanitarian, and rear-area enforcement, then enters an armed diplomatic pause.	CENTCOM hits Bushehr, Chah Bahar, Jask, Konarak, Abu Musa, Bandar Abbas, Greater Tunb, and Qeshm; the toll is withdrawn but the blockade remains; Belma is disabled and 17 commercial vessels are redirected; AP reports bridges toward Bandar Abbas, the Chabahar tower, a tanker abandoned in the strait, fires or damage on vessels along the Oman route and off the UAE coast, retaliation across Qatar, Kuwait, Jordan, Bahrain, Iraq, and Saudi Arabia, Pickaxe Mountain never inspected by the IAEA, IMO rises to 62 confirmed incidents as of July 27 with 17 seafarer fatalities, the Houthis hit Saudi shipping at Bab el-Mandeb, Yanbu and Jizan enter the Houthi-Saudi exchange, the UN/OHCHR call covers at least 6,000 seafarers stranded on about 400 vessels, AP records three days without new U.S. or Iranian attacks, Qatar-Pakistan mediation progress, Iran-Oman talks on the strait mechanism, and commercial Hormuz traffic at a three-week low, while Tehran keeps the strait closed and Brent falls to USD 84.34 in MarketWatch’s July 28 intraday continuous quote.
July 28-30	The pause breaks in the rear area, then returns to IRGC targets in Iran, while Hormuz shows partial capacity without normalization.	CENTCOM and Saudi forces strike Tehran-aligned groups in Iraq after more than 30 IRGC-directed drone attacks; CENTCOM places a new heavy wave against dozens of IRGC targets in Iran at 10 p.m. ET on July 29 after attempted missile attacks on U.S. forces; AP reports five Iranian missiles intercepted by Jordan with no casualties and no imminent direct diplomacy; Iran rejects Oman’s proposed equal split of routes; the IRGC claims three tankers were struck or stopped, but IMO remains at 62 confirmed incidents as of July 27; Al Areesh exits Hormuz as the first visible QatarEnergy LNG transit since July 11 and MarketWatch records Brent continuous at USD 87.29 and WTI continuous at USD 83.66 in July 30 observations.
July 31-August 1	The route remains passable but warning-heavy, while the rear area widens again.	AP records no overnight U.S. strike on July 31 but new strike options under White House review, then new Trump threats on August 1; UKMTO issues two warnings near Lima and Khasab for incidents reported on July 31, one with tanker engine-room damage and one nearby explosion without declared damage; Kuwait denounces Iranian attacks against vital installations and civilian vehicles on Boubyan Island; the Houthis force eight Saudi ships away from Bab el-Mandeb; Tehran keeps Hormuz as leverage and denies direct requests to Washington; MarketWatch records Brent continuous at USD 91.04 and WTI continuous at USD 86.80 in July 31 delayed quotes.
August 2	The pause becomes negotiating space again, but not verified reopening.	AP records Trump’s decision to hold off new U.S. strikes because regional allies had reached the parameters of a deal; the proposal would include reopening Hormuz, halting regional attacks, ending the naval blockade, and allowing Iranian oil exports, but no deal has been reached. Tehran says Hormuz will not return to its pre-February 28 status, and Al Jazeera records ships stuck in the northern corridor, with passage conditioned on Iranian authorization even for the southern corridor in Omani waters. IAEA, EIA, and CENTCOM add no new public numbers.
August 3	The negotiating window produces administered transits and repricing, not normalization.	AP records the announcement of U.S.-Iran talks beginning Monday, Iran’s denial of direct talks, and nine vessels transiting Sunday according to Kpler, mostly through an Iranian-controlled route; Al Jazeera places the Oman-Iran route in final stages; IMO rises to 63 confirmed incidents, including GASLOG SHANGHAI damaged near Lima on July 31; MarketWatch shows Brent continuous at USD 83.62 and WTI continuous at USD 80.40 in the August 3 delayed quote, while settlement, IAEA, EIA, and CENTCOM remain anchored to previous releases.
August 4	The Iran-Oman draft makes the transit regime more legible, but a new warning blocks a normalization reading.	AP records progress on a formula with entry through an Iranian-controlled route, exit through an Oman-controlled route, service fees, and a link to lifting the U.S. blockade on Iranian ports; Al Jazeera confirms positive talks on safe lanes and Tehran denies direct talks with Washington; AP cites UKMTO on a new cargo ship hit by an unknown projectile off Oman, with damage reported by Ambrey; at the time of verification IMO remains at 63 confirmed incidents and MarketWatch shows Brent continuous at USD 79.51 and WTI continuous at USD 75.94 in the August 4 delayed quote.
August 5	The draft becomes nearly signable, but the operational ledger worsens.	AP records an Iran-Oman draft finalized pending final Iranian approval, with Trump pointing to a possible deal within 24-48 hours and contesting tolls; Al Jazeera describes residual issues over lane control, service fees, and transit approval, with eight vessels passing Monday compared with roughly 130 per day before the war; IMO rises to 64 confirmed incidents including MINOAN PIONEER, damaged on August 3 near Khasab with one seafarer missing; CENTCOM and IAEA add no relevant releases, EIA publishes the August 5 report, and MarketWatch shows Brent continuous at USD 80.07 and WTI continuous at USD 75.27 in the August 5 delayed quote.
August 6-10	The draft becomes a framework, then hardens into broader political conditions while traffic remains administered.	Al Jazeera records an Iran-Oman framework with final approval still pending in Tehran; AP places the bottleneck around possible U.S. compromise on control of the strait, the blockade on Iranian ports, and the absence of permits, tolls, or impediments on temporary routes, then attributes broader Iranian conditions to the Supreme National Security Council on a permanent end to war, military withdrawal, the port blockade, compensation, sanctions, and frozen assets; AP/Lloyd’s List puts traffic at 84 transits during the week of July 27-August 2, up from 45 the previous week but still below more than 700 in a typical pre-crisis week; UKMTO Warning 108-26 records an August 8 projectile strike 18 nautical miles east of Khasab and ADNOC reports three of its own vessels hit during the week, but IMO remains at 64 confirmed incidents as of August 4; on August 10 Al Jazeera records only partial negotiation according to Trump and no direct talks according to Araghchi, with Hormuz still tied to the end of the naval blockade; IAEA, CENTCOM, and EIA add no new relevant release; MarketWatch shows Brent continuous at USD 84.14 and WTI continuous at USD 78.44 in intraday delayed quotes, while AP records the August 9 Houthi claim against the Aramco refinery in Jizan with no reported casualties.
August 11-19	Control legitimacy meets operating data, confirmed incidents, and diplomatic deadlines.	Trump claims U.S. control of the strait and seeks compensation from Iran; Tehran keeps its conditions on the blockade, sanctions, frozen assets, compensation, and an end to threats, then says on August 13 that no transit can occur without Iranian permission; on August 15 Tehran has not decided to return to talks with Washington, distinguishes messages via Qatar and Pakistan from actual negotiations, works with Oman on a temporary technical route, and restates Iranian control of Hormuz; according to CENTCOM as reported by AP and Al Jazeera, a U.S. helicopter hits the Panama-flagged Vela Nova cargo ship as it tries to sail toward an Iranian port; CENTCOM then raises the blockade tally to 62 redirected commercial vessels, three disabled vessels, and two boarded vessels; the August 14 UKMTO overview keeps IMO traffic separation arrangements suspended, measures traffic about 90% below pre-conflict baselines, and counts 20 UKMTO projectile strike incidents since July 6; Kpler counts eight Tuesday transits, then Reuters/Kpler places traffic near standstill with nine Thursday crossings, five Wednesday crossings, two visible Friday crossings, and no observable oil cargo; two ADNOC vessels are hit by drones in Hormuz on August 14 with no injuries, then UAE/WAM attributes a new attack against an ADNOC-affiliated vessel to Iran on August 15 and AP records a bulk carrier hit in the hull according to UKMTO; the Houthi attack on Tihamah kills six people in Bab el-Mandeb and AP records a new Houthi attack on Mokha port with three deaths and suspended operations; AP also records spills off Qeshm and Oman, Houthi-government clashes in Taiz, a new drone strike against Jizan, and George Washington rotating toward the Middle East while Lincoln exceeds 240 uninterrupted days at sea; CENTCOM announces Task Force Falcon Strike; the 60-day MoU term expires on August 17 without extension and on August 18 Washington says no talks are scheduled, while Tehran keeps reopening tied to U.S. commitments and the Iran-Oman draft remains contested; IMO rises to 66 confirmed incidents and 18 fatalities including MINOAN DIGNITY, damaged in the Strait of Hormuz on August 17 with one seafarer fatality and no pollution; AP/Kpler records 95 crossings in the previous week, down 19.5%, only three on Sunday, traffic on the Iran-designated route, and no crossings on Omani routes; AP also records a projectile strike off Oman with engine-room damage and a casualty, and the UAE says two missiles fell into the sea without damage or injuries, an attribution Tehran disputes; the UKMTO recent incidents page checked on August 19 remains at 0 current reports; IAEA adds no new public access and remains at the February and June Board Reports and the June 10 GOV/2026/40 resolution; the August 11 EIA STEO estimates Hormuz at 4.9 mb/d in 2Q26, shut-ins at 5.5 mb/d in July, and 3Q26 Brent around USD 85/b; EIA WPSR remains at the August 12 release for the week ending August 7; the August IEA OMR cuts 2026 demand and 3Q26 supply; MarketWatch puts the August 18 settlement at USD 91.02 for Brent continuous and USD 84.06 for WTI continuous.
August 20-23	Pressure shifts from lanes to the combination of AIS opacity, sanctions, naval capital, ownership claims, and logistical bypasses.	Al Jazeera/Kpler reclassifies August 1-19 transits: 236 total crossings, 148 dark or unclassified, 83 on the Iran-designated route, three on Omani routes, and two on the pre-war central channel; for the 112 oil-and-gas vessels, 89 are dark or unclassified, 21 use the Iranian route, and two use the Omani route. IMO, updated to August 21 and checked on August 23, remains at 68 confirmed incidents and 19 fatalities, with AL WATAN in the Strait of Hormuz on August 14 with no pollution, alongside RIYAN STAR and MINOAN DIGNITY; UKMTO recent incidents remains at 0 current reports in the August 23 check, while the VRA overview for the week ending August 21 keeps risk SEVERE in Hormuz, IMO traffic separation procedures suspended, and 23 UKMTO-reported projectile strike incidents since July 6. AP records George Washington’s arrival in the Arabian Sea to relieve Lincoln and Iran’s response to new U.S. economic threats; Al Jazeera records the promise of tougher secondary sanctions and Chinese criticism. On August 22 Pezeshkian speaks of ending the war from a position of strength, Oman and Iran discuss conditions for renewed dialogue, and Trump claims Hormuz as American territory; the CENTCOM tally carried by the Guardian brings the blockade to 68 redirected commercial vessels, three disabled, and two boarded. AP adds Rezaei, Iranian threats against neighbors that join the U.S. squeeze, Egypt’s attempt to revive the U.S.-Iran channel, unquantified facilitation for some Iraqi tankers, and a French-Saudi agenda around Omani ports, pipelines, and rail bypasses. IAEA remains without new public access; EIA keeps WPSR at the August 19 release for the week ending August 14; MarketWatch puts the August 21 settlement at USD 92.67 for Brent continuous and USD 87.06 for WTI continuous.

The sequence keeps the same grammar: every ceasefire is measured by the corridor, every price anticipates a trust that ships must then confirm, and every broker matters as long as it can reduce the cost of the route rather than merely transmit incompatible messages.

Observations, not indices

The previous 0–100 conflict and Italy-risk scores have been removed. Their weights mixed military events, maritime conditions, prices, diplomacy, and Italian exposure without a dataset capable of identifying the coefficients. That made the result responsive to the analyst’s choices and, in the case of energy prices, partly circular.

The verified series still matter, but their roles are now explicit. Incidents, transits, waiting times, diversions, insurance conditions, sanctions, and IAEA access inform the operational state. Brent and TTF are downstream outcomes or auxiliary sensors. Italy’s import volumes, exchange rate, freight, sectoral energy exposure, inflation, and output belong to the transmission stage.

The prices recorded in the chronology — including Brent at USD 76.01 on July 10, USD 84.95 on July 15, USD 84.23 on July 16, USD 88.10 on July 17, USD 89.22 at the July 20 settlement, USD 91.01 at the July 21 settlement, USD 94.07 at the July 22 settlement, and USD 100.69 at the July 23 settlement — remain observations, not a loss estimate. MarketWatch shows the Brent continuous contract at USD 92.82 at 22:59 BST on July 24, USD 85.89 at 12:07 BST on July 27, USD 84.34 at 08:13 BST on July 28, USD 84.82 at 07:11 BST on July 29, USD 87.29 at 05:14 BST on July 30, USD 91.04 at 22:59 BST on July 31, with a USD 87.93 settlement, USD 83.62 at 18:26 BST on August 3, USD 79.51 at 17:21 BST on August 4, USD 80.07 at 14:25 BST on August 5, with August 4 settlement at USD 79.36, USD 82.38 at 22:59 BST on August 7, with settlement at USD 83.55, and USD 84.14 at 5:43 BST on August 10, USD 87.72 at 4:32 BST on August 11, matching the August 10 settlement, USD 89.50 at 02:05 BST on August 12, and USD 88.55 at 22:59 BST on August 12, with an August 12 settlement at USD 88.98; then USD 86.93 at 21:20 BST on August 13, USD 88.82 at 22:58 BST on August 14, with an August 14 settlement at USD 88.52, and USD 91.53 at 9:53 BST on August 19, with an August 18 settlement at USD 91.02, then USD 93.60 at 5:32 BST on August 21, with an August 20 settlement at USD 93.78, and USD 92.22 at 22:59 BST on August 21, with an August 21 settlement at USD 92.67; WTI continuous moves from USD 83.05 at 06:29 EDT on July 27 to USD 80.26 at 04:35 EDT on July 28, USD 82.05 at 01:36 EDT on July 29, USD 83.66 at 23:49 EDT on July 29, USD 86.80 at 16:59 EDT on July 31, with a USD 84.67 settlement, USD 80.40 at 14:43 EDT on August 3, USD 75.94 at 12:23 EDT on August 4, USD 75.27 at 11:12 EDT on August 5, USD 77.08 at 16:59 EDT on August 7, with settlement at USD 78.18, USD 78.44 at 1:21 EDT on August 10, USD 82.25 at 23:27 EDT on August 10, with settlement at USD 82.13, USD 83.83 at 22:01 EDT on August 11, and USD 82.77 at 18:15 EDT on August 12, with an August 12 settlement at USD 83.27, then USD 81.21 at 16:59 EDT on August 13, USD 82.40 at 16:59 EDT on August 14, with an August 14 settlement at USD 82.40, and USD 84.56 at 4:17 EDT on August 19, with an August 18 settlement at USD 84.06, then USD 86.20 at 1:06 EDT on August 21, with an August 20 settlement at USD 86.83, and USD 86.64 at 16:59 EDT on August 21, with an August 21 settlement at USD 87.06. These values document repricing, a correction, and a rebound in a continuous series that is also sensitive to contract rollover; EIA WPSR is updated to the August 19 release for the week ending August 14 and adds the inventory confounder in the same window, while the August 11 EIA STEO makes the expected scarcity of physical flows through Hormuz explicit and the August IEA OMR quantifies global stock drawdown and the demand revision. Global demand, OPEC+ decisions, inventories, the dollar, insurance, and liquidity must be controlled for before attributing the move to Hormuz.

Observed series Brent during the crisis

Price in US dollars per barrel. The line shows observations, not a closure probability or an estimate of Italy's economic loss.

Latest observation $92.67 August 21, 2026 · MarketWatch settlement

EIA spot · through June 29 Verified market benchmarks · July 10-August 21

Operational dates help organize the sequence, but proximity to a price move does not by itself identify a Hormuz shock. Global demand, OPEC+, inventories, the dollar, and liquidity remain confounders to control for.

Operational scenarios

Each scenario is now a conditional and falsifiable hypothesis. It is accepted or rejected by transits, capacity, waiting times, insurance, physical losses, and verified attacks, not by a composite score.

Prolonged friction. Strikes and retaliation remain below a generalized closure but keep Hormuz expensive, brokers under pressure, and negotiations incomplete. The hypothesis gains support if transit counts hold while waiting times, war-risk premiums, and diversions remain persistently above their ordinary baseline.

Route war. This is the clearest form of the July round: Tehran pushes toward a route recognized by the IRGC, Oman and the IMO keep an alternative framework open, the United States responds with strikes and sanctions, and commercial operators decide on the basis of crews, coverage, and waiting times. After July 14, the form becomes harder: the US blockade is selective against Iranian ports, but its enforcement area intersects the same geography in which Tehran claims control, threatens routes, and tries to convert passage into political leverage. The corridor remains passable but expensive, contested, and reversible.

Contained de-escalation. The ceasefire does not have to become peace to produce real effects, but the August 2-5 window has to become observable. A few technical checks, a less opaque IAEA channel, less selective maritime routes, a more stable role for Oman and Qatar, and a verifiable reduction in attacks on Gulf host nodes would be enough. The early signal is not a statement, but the return of crews, insurance coverage, and ordinary transits.

Regional widening. The jump in scale does not require a single decision. It can emerge from the sum of three failure modes: Lebanon reactivating the Israeli front, Bab el-Mandeb duplicating the Hormuz problem, and a Gulf node hit severely enough to force a broader coalition response. This is the scenario in which the network stops absorbing shocks and begins propagating them.

Data that distinguish the scenarios

Here the model becomes operational: a news item matters when it changes an observable input, a latent-state probability, or a physical-flow forecast, not when it merely lengthens the chronology.

Actual Hormuz transits: number of ships, routes, waiting times, incidents, refusals to pass, and insurance conditions.
Seafarers and IMO incidents: evacuations, crews detained or stranded, the count of confirmed incidents, and companies’ ability to find personnel willing to transit.
Ceasefire status: not only statements, but observed strikes, retaliation, bases involved, and naval rules of engagement.
Licenses and sanctions: OFAC revocations, wind-down windows, bans on new cargoes, and signs that the conflict is moving from the sea into the legal pricing of oil.
IAEA verification: access, technical continuity, declared material, and disputes over the mandate.
Regional brokers: Oman, Qatar, Pakistan, and the UN matter if they shorten the channel; they matter less if they become only megaphones for incompatible messages.
Gulf host nodes: Kuwait, Bahrain, Qatar, the UAE, and Saudi Arabia are the best sensors of the rear area’s transformation into a front.
Markets and the Italian economy: oil, LNG, war-risk premiums, shipping, and signals from energy-intensive industry must be read together, not as isolated indicators.

A news item changes the reading of the crisis when it updates at least one of these measurements. The chronology reconstructs what happened; a predictive claim begins only when that information changes a probability distribution and survives an out-of-sample test.