A Monte Carlo and decision-tree risk analysis of a U.S. plan to seize Kharg Island and its impact on oil prices, war risk, and global escalation.
 |
| Cover Image Attribute: The satellite image of Kharg island / Source: Planet Labs |
Senior officials have confirmed no final decision has been made, but Trump has signaled resolve: “We can take out the island anytime we want.” Preceding U.S. strikes on March 13, 2026, already neutralized key military assets on Kharg—air defenses, naval bases, missile bunkers—while sparing export infrastructure to preserve leverage. Additional Marine expeditionary units are en route, bringing total amphibious forces potentially to 7,500 or more. The strategic calculus is clear: reopen Hormuz or face direct control of Iran’s export chokepoint. Yet analysts warn the move carries profound risks of backfire.
To evaluate this decision rigorously, we apply a probabilistic decision-tree risk model combined with Monte Carlo simulation (10,000 iterations) and standard economic elasticity frameworks. Define four mutually exclusive outcomes with assigned probabilities derived from military feasibility assessments, historical precedent (e.g., 1980s Tanker War), and expert warnings in the referenced reporting:
- Success (U.S. seizes Kharg and forces Hormuz reopening): ps = 0.25
- Limited escalation (regional proxy conflict, partial disruption): pl = 0.35
- Full military escalation (U.S.–Iran war, Hormuz closure): pe = 0.25
- Internationalization (global powers drawn in): pi = 0.15
These probabilities sum to 1 and reflect conservative estimates: low success odds due to Iran’s alternative export routes (pipelines, smaller ports) and high escalation risk from IRGC retaliation capabilities. Associated Brent crude price outcomes (baseline current ~$110/bbl) are $80, $130, $160, and $220 respectively. The expected oil price is calculated as:
EV[P] = ∑ pk ⋅ Pk = (0.25 × 80) + (0.35 × 130) + (0.25 × 160) + (0.15 × 220) = $138.5
Monte Carlo simulation confirms: mean $137.63/bbl (the difference is pure statistical noise), median $130, standard deviation $44.40, with 38.9% probability of exceeding $150/bbl and 14.5% chance of surpassing $200/bbl. Global economic extra annual cost (at $36.5 billion per $1/bbl above $100 baseline, scaled to 100 million bpd world consumption) averages $1.37 trillion—equivalent to roughly 1.3% of global GDP in a sustained shock. Short-run oil demand elasticity (ϵd≈−0.15, per macroeconomic models) implies that a 20% Hormuz disruption would drive prices up ~133% via:
ΔP/P = −(ΔQ/Q)/ϵd
This framework quantifies the gamble: success yields net global benefit; failure triggers cascading losses.
We now analyze each scenario.
Success Scenario: Control of Kharg Compels Iran to Reopen Hormuz
In the success case (ps = 0.25), U.S. forces secure Kharg after preliminary strikes degrade Iranian coastal defenses. Marines establish a perimeter, blockading tanker loading while leaving export pipes intact as bargaining leverage. Iran, facing total severance of 90% of its export revenue (~$70–80 billion annually at current prices), faces internal pressure from economic collapse and potential regime instability. Tehran capitulates within weeks, reopening Hormuz to international shipping under U.S.-escorted convoys.
Mathematically, the payoff is modeled as avoided disruption cost. With Hormuz flows restored (21 million bpd), global supply normalizes. Using the elasticity equation above, prices fall to $80/bbl—a 27% drop from baseline. Annual global savings: (110−80) × 36.5 × 109 ≈ $1.1 trillion. U.S. strategic gains include restored energy security for allies, weakened IRGC funding (cutting its war chest by billions monthly), and a demonstration of resolve that deters future chokepoint closures. Monte Carlo subsets for this scenario show 100% of success simulations yield prices below $100/bbl within six months, with GDP uplift of 0.4–0.8% globally (per DSGE models linking oil shocks to output).
Militarily, holding Kharg requires ~5,000–7,500 troops with air and naval support. Casualty risk is minimized post-degradation (Iranian missile sites neutralized). Proxy attacks diminish as IRGC prioritizes survival over retaliation. Long-term, control provides a forward base for monitoring Iranian mainland activity, potentially accelerating regime change dynamics Trump has referenced. Net expected value for this scenario alone: positive $275 billion annualized (probability-weighted savings minus occupation costs estimated at $20–30 billion/year for logistics and security). Success restores pre-crisis stability faster than diplomatic alternatives, validating the “shot across the bow” airstrikes already conducted.
Critically, this outcome hinges on precise execution: sparing oil infrastructure prevents total Iranian economic desperation that could harden resistance. Historical parallels (1980s war damage to Kharg took years to rebuild) underscore leverage, but only if U.S. forces avoid over-destruction that invites insurgency.
Backfire 1: Immediate Military Escalation into U.S.–Iran Regional War
With probability 0.60 conditional on failure (pe + pi weighted), Iran responds asymmetrically rather than concede. Likely actions include ballistic missile barrages on U.S. Gulf bases (e.g., Al Udeid, Al Dhafra), swarming fast-boat and mine attacks on U.S. naval assets, and proxy activations via Hezbollah, Houthis, and Iraqi militias striking Saudi, Emirati, and Israeli targets. This rapidly escalates into a full regional war, with Hormuz remaining closed or heavily contested.
Risk modeling employs a Poisson process for attack frequency: λ = 15 major incidents per month (based on Iran’s documented drone/missile capacity and proxy networks). Probability of at least one U.S. base strike in the first 30 days is 1−e−λ ≈ 99.99%. Casualty modeling (binomial distribution, n = 50 potential strikes, success probability per strike 0.20 from defensive intercepts) yields expected U.S./allied deaths of 200–500 in month one, scaling with escalation intensity.
Oil impact compounds: even partial Hormuz interdiction (10% effective closure) triggers ΔP ≈ 67%
via elasticity, pushing prices to $160–180/bbl in this scenario. Simulation shows 89% of escalation scenarios exceed $150/bbl. Broader war costs—U.S. munitions expenditure ($5–10 billion/month), carrier group deployments, and ally base repairs—add $150–300 billion annually. Global GDP contraction: 0.5–1.2% (Vanguard/NIESR models for $125+/bbl shocks inducing recession in Europe/Japan).
Iran’s retaliation playbook is well-rehearsed: surviving initial strikes, it targets Gulf energy infrastructure (desalination plants, ports) as warned by officials. This creates a feedback loop—each U.S. reinforcement raises Iranian resolve, modeled as a game-theoretic escalation ladder where payoff matrix favors asymmetric persistence (Iran loses less per engagement). Expected value for this branch: –$850 billion annualized (price shock + direct costs + 0.8% global GDP loss), dwarfing any territorial gains on Kharg. The operation risks U.S. quagmire, echoing analyst warnings of Vietnam-like entrapment.
Backfire 2: Global Oil Price Shock and Inflationary Spiral
Independent of pure military escalation, seizing Kharg triggers immediate supply shock. Removing 2 million bpd (2% global) directly, plus insurance spikes and shipping aversion, effectively idles 5–10% of flows initially. Full Hormuz risk (20% trade) amplifies via elasticity: 2% loss yields ΔP ≈ 13% baseline, but panic and hoarding multiply to 30–50% spikes.
Monte Carlo isolates this channel: mean price $145/bbl across shock-heavy paths, with 52% probability above $150. At $150/bbl sustained, models project:
- U.S. headline inflation +0.8–1.2 percentage points (New York Fed DSGE estimates for comparable shocks).
- Euro-area GDP –1.0% (recession threshold at $125+).
- Global extra oil import bill: $1.8 trillion/year.
Insurance rates for Gulf tankers could surge 300–500% (historical precedent), adding $20–30/bbl logistics premium. Inflationary transmission: core prices rise 0.2–0.4% via second-round wage/price effects. Emerging markets face debt crises as import bills balloon.
Mathematically, expected shock cost is
E[C] = pshock × (ΔP × 36.5 × 109 + GDP multiplier × 0.003 × World GDP)
With pshock = 0.75 (weighted failure), this equals ~$1.1 trillion annual drag. Unlike success, where prices normalize quickly, this scenario sustains elevated volatility (standard deviation $35+/bbl in simulations) for 6–18 months until alternative routes (Iraq pipelines, Chinese offloading) scale—yet these handle only ~30–40% of lost volume. The shock undermines Trump’s domestic agenda, risking midterm backlash from $5+/gallon gasoline and renewed inflation.
Backfire 3: Internationalization into Major Global Confrontation
Failure draws third parties, shifting from bilateral to multipolar crisis. China (60%+ of Iranian oil buyer) faces direct supply threat; Russia provides diplomatic cover and indirect arms/logistics. European/Gulf states are pulled in via refugee flows, energy rationing, and alliance obligations.
Game-theoretic modeling: a 3-player payoff matrix (U.S., Iran + Russia/China bloc, Europe/Gulf) where internationalization yields negative-sum outcomes. Nash equilibrium favors escalation—China diverts tankers and pressures OPEC+, Russia vetoes UN resolutions. Probability-weighted internationalization scenario (pi = 0.15, conditional 0.25 on failure) produces $220/bbl prices: 20% supply loss drives ΔP ≈ 133%, plus multiplier effects from coordinated retaliation.
Simulation: 100% of internationalization runs exceed $200/bbl; expected global GDP loss 2.5–3.5% (compounding oil shock with financial contagion). China’s alternative sourcing strains its economy (–0.7% GDP), prompting export retaliation or South China Sea tensions. Russia gains by selling discounted oil, prolonging conflict. Expected value: –$2.1 trillion annualized (price + geopolitical risk premium).
Broader effects include nuclear proliferation signals (Iran accelerates enrichment) and alliance fractures. Internationalization probability rises with perceived U.S. overreach, modeled via Bayesian update: prior 0.15 posterior 0.40 if initial strikes fail to deter proxies.
Overall Risk Assessment and Decision Implications
Aggregating via the decision tree, total expected oil price is $138.5/bbl with $1.37 trillion annual global extra cost—net negative versus status-quo diplomacy. Success probability must exceed 0.45 for positive EV, far above realistic estimates given Iran’s resilience and alternative routes. Sensitivity analysis (varying ps ± 0.10) confirms robustness: even at 0.35 success, expected loss exceeds $800 billion.
The Kharg operation offers asymmetric upside if executed flawlessly but carries fat-tail downside risks—regional war, recessionary oil shocks, and global confrontation—quantified here at 75% aggregate failure probability. Mathematical modeling reveals the plan as a low-probability, high-reward gamble ill-suited to current force posture and market vulnerabilities. Prudent alternatives—convoy escorts post-degradation or targeted sanctions—yield higher expected utility with lower variance. Trump’s fixation on Hormuz reopening is understandable, yet the data-driven risk calculus underscores why targeting Kharg could transform a manageable crisis into a generational strategic setback.
Omission of Israel's Role in the Analysis: The Unpredictability Factor
Israel’s role as perhaps the single largest aggravating factor in the U.S. risk calculus for seizing or blockading Kharg Island was deliberately omitted from the original probabilistic decision-tree modeling and expected-value calculations, primarily due to its extreme unpredictability, its very low likelihood of being part of a U.S. Marine ground offensive, and the resulting difficulty of assigning reliable, quantifiable probabilities. The conflict originated as a joint U.S.–Israeli operation on February 28, 2026, with coordinated strikes targeting Iranian military and nuclear assets and killing Supreme Leader Ali Khamenei. However, Israel’s subsequent independent actions—most notably the March 18 airstrikes on the South Pars gas field, which damaged approximately 12% of Iran’s gas production and halted exports—have diverged from U.S. priorities, which remain focused on reopening the Strait of Hormuz and stabilizing oil prices.
This divergence reflects fundamentally different strategic priorities. U.S. objectives center on reopening Hormuz and stabilizing global oil prices, a domestic political and economic priority amid oil prices above $110 per barrel and persistent inflation concerns. Israel, by contrast, is pursuing broader strategic degradation of Iran’s capabilities, including targeting nuclear infrastructure and IRGC-linked energy revenue sources through strikes such as the March 18 attack on South Pars. A U.S. ground operation on Kharg Island aligns with Washington’s coercive leverage strategy—controlling export infrastructure to force Hormuz reopening—but does not fully align with Israel’s broader objective of long-term Iranian capacity degradation. This creates the risk of alliance friction, particularly if Israeli forces were to pursue independent actions affecting oil infrastructure on the island, such as destroying facilities that the U.S. has deliberately spared to preserve economic leverage.
Netanyahu’s pursuit of broader regime-weakening objectives therefore creates path-dependent escalation dynamics: each Israeli strike hardens Iranian resolve, activates proxy networks more aggressively against both U.S. and Israeli targets, and raises the baseline level of escalation in ways that do not fit clean probabilistic scenario modeling. This unpredictability—especially sudden, uncoordinated strikes on energy infrastructure that increase global oil and gas price volatility beyond U.S. control—effectively makes Israel a “wild card” variable that inflates tail risks and undermines assumptions of separable U.S.–Iran conflict dynamics.
Incorporating this variable into the model would require subjective probability multipliers or additional high-variance sub-scenarios—for example, a 60–70% probability of further independent Israeli actions during a Kharg operation—which would significantly widen outcome distributions and reduce model stability. The original simplified model therefore prioritized tractable, U.S.-centric scenarios to avoid compounding uncertainty into analytically uninformative outputs. As a result, the baseline expected oil price estimate of $138.5 per barrel and the 75% aggregate failure probability likely understate the true downside risk by an estimated 20–40%, because Israeli actions narrow the success pathway while making escalation scenarios—regional war and sustained oil prices above $150–200 per barrel—more probable and more severe.
The above risk modeling—built on a probabilistic decision-tree framework, Monte Carlo simulation (10,000 iterations), and short-run oil demand elasticity assumptions—provides a structured, quantitative lens for evaluating the Kharg Island operation. However, like all formal risk models applied to geopolitical-military crises, it carries significant inherent limitations that reduce its predictive power, inflate uncertainty, and risk misleading decision-makers. Below, we ourselves outline the key constraints, drawing from established critiques in conflict forecasting, strategic risk analysis, and energy market modeling.
1. Subjectivity and Arbitrary Probability Assignments
The model's core relies on assigned branch probabilities (e.g., success ps = 0.25, escalation pe = 0.25, internationalization pi = 0.15). These figures are expert-derived estimates informed by historical analogies (e.g., 1980s Tanker War) and current reporting, but they remain inherently subjective. In geopolitical contexts, probabilities are not empirically observable frequencies; they reflect judgments about adversary intentions, resolve, miscalculation risks, and unknown unknowns.
Critiques of probabilistic models in violent conflict and nuclear risk highlight that such assignments often overstate precision. Autoregressive models using only lagged violence data frequently outperform covariate-rich structural models, suggesting that deep structural factors (regime stability, proxy dynamics, leadership psychology) add little predictive value and introduce bias. When probabilities are misspecified—even modestly—the expected value swings dramatically. Sensitivity tests in the original model show that raising success probability to just 0.35 flips the net outcome positive, underscoring fragility to input assumptions.
Moreover, the mutually exclusive branches oversimplify real-world path dependence: escalation and internationalization are not cleanly separable; proxy attacks can rapidly internationalize via Chinese or Russian responses, creating correlated risks the tree ignores.
2. Inability to Capture Endogenous Dynamics and Adaptive Behavior
Decision trees and Monte Carlo simulations assume relatively static scenarios where actors follow pre-defined paths. In reality, U.S.–Iran interactions are endogenous: Iran's responses adapt to U.S. actions (e.g., initial restraint after March 13 strikes, then rapid proxy mobilization if invasion proceeds), while U.S. escalation thresholds shift based on casualties or ally pressure.
Models struggle with this feedback. Adversaries are not passive; they employ asymmetric tactics (mines, swarms, proxies) precisely to exploit U.S. vulnerabilities, creating non-linear escalation ladders absent from the tree. Historical cases (e.g., Iran-Iraq War, Israel-Hamas dynamics) show that conflict trajectories defy probabilistic forecasting because actors learn, bluff, or miscalculate in ways inductive models cannot anticipate. Deep learning or Monte Carlo approaches, while powerful for sampling uncertainty, remain rooted in inductive logic—pattern-matching from past data—yet warfare's "fog" (incomplete information, friction, chance) ensures novel behaviors emerge.
The model also underweights human elements: leadership psychology, domestic politics (e.g., IRGC hardliners vs. pragmatists; U.S. midterm pressures), and misperception risks are treated as noise rather than drivers.
3. Limitations of Oil Price Elasticity in Sudden, Geopolitical Shocks
The elasticity-based price projections (e.g., ΔP/P ≈ –(ΔQ/Q)/ϵd with ϵd ≈ –0.15) assume smooth, proportional responses to supply disruptions. Short-run demand elasticity is indeed low, but sudden Hormuz-style shocks trigger non-linear effects:
- Panic and precautionary hoarding: Markets price in perceived long-term loss of access to Gulf reserves (55% of global proven), not just barrels removed. Fear prevents inventory releases, amplifying spikes beyond elasticity predictions.
- Behavioral multipliers: Speculative positioning, insurance surges (300–500%), and rerouting costs add premiums uncorrelated with physical barrels lost.
- Asymmetric responses: Historical disruptions (1973 embargo, 1990 Gulf War) show prices overshooting proportional models due to perception of enduring supply risk, even when physical shortfalls are modest or temporary.
- Endogeneity of uncertainty: Oil price volatility responds to macroeconomic disaster risk (e.g., recession fears) as much as geopolitical events, invalidating assumptions of exogenous supply shocks.
Elasticity models fit gradual adjustments well but fail in crisis regimes with sparse historical data points, where regressions fit noise rather than signal. Monte Carlo runs inherit this flaw, producing tight confidence intervals around biased central estimates.
4. Over-Reliance on Historical Analogies and Data Limitations
The model extrapolates from past crises, yet current conditions differ markedly: Iran's export diversification (pipelines to China, ghost tankers), global spare capacity (mostly non-Gulf), U.S. energy independence, and proxy networks' evolution reduce direct parallels to 1980s or 1970s events. Conflict forecasting research shows point predictions without uncertainty bands often underperform simple heuristics, with hidden uncertainty guaranteeing errors.
Data on Iranian capabilities (missile accuracy, proxy coordination, mine-laying speed) remains classified or uncertain, while U.S. degradation effectiveness post-strikes is speculative. Monte Carlo amplifies rather than resolves this by sampling around incomplete distributions.
5. Tail-Risk Underestimation and Black-Swan Blind Spots
Fat-tail events—nuclear threshold crossing, accidental escalation involving Russia/China, regime collapse in Tehran triggering civil war, or cascading financial contagion—are underrepresented. The assigned 0.15 internationalization probability may capture baseline risk but misses low-probability, high-impact paths where small missteps cascade globally.
Monte Carlo excels at variance within defined scenarios but cannot simulate structural breaks or emergent phenomena outside the input distribution.
Implications for Decision-Making
These limitations mean the model's $138.5/bbl expected price and $1.37 trillion annualized cost carry wide, unquantifiable error bands—potentially rendering the net-negative verdict overly confident or, conversely, understating catastrophe risk. Quantitative outputs risk false precision, encouraging over-reliance while sidelining qualitative judgment, historical nuance, and real-time adaptation.
In high-stakes geopolitical gambits like Kharg, models best serve as sensitivity tools or red-teaming aids, not prescriptive forecasts. Decision-makers should weight them against:
- Scenario planning emphasizing worst-case persistence and miscalculation.
- Adversarial risk analysis incorporating opponent Bayesian updating.
- Human oversight to capture friction, morality, and context absent from algorithms.
Ultimately, while the modeling illuminates trade-offs, its constraints remind us that Clausewitzian uncertainty—chance, friction, and the fog of war—defies full quantification. The Kharg operation remains a gamble where mathematics clarifies stakes but cannot eliminate the fog.
Apple News, Google News, Feedly, Flipboard, and WhatsApp Channel
