Will the US Invade Iran in 2026? The Odds, and Kharg

Published 27 September 2026 · Prices frozen 27 September 2026, 16:20 UTC

Every page currently ranking for will the US invade Iran quotes a number that is at least eleven days old and rounds it to the nearest point: 14%, 17%, one of them 55–65% on a Kalshi contract that does not exist. The live figure is 15.5¢ on $69.76m of traded volume, and the round number is the least interesting thing about it, because the same exchange also prices the specific objective every reported ground plan this year was built around — taking Kharg Island — at 5.5¢. We read both venues in the same pass, pulled the full lifetime daily history on every contract in the cluster, swept 40 pages of Kalshi’s open events to establish what it does and does not list, and put the two resolution texts side by side. They are not asking the same question, the ten points between them are the answer to a better one, and by year-end the market is quietly pricing a resumed American war rather than an invasion.

Where the US-Iran invasion odds stand today

Four Polymarket contracts cover the military question directly, and a fifth covers where it ends. All of them settle on 31 December except the near-dated Kharg rung, which settles in three days.

The thing to notice is not the level but the ordering. The broadest, cheapest-to-satisfy contract in the set — an offensive begun against any portion of Iranian territory — is priced at nearly three times the contract on actually holding one island, and at almost six times the contract on Congress voting for it. That is not a contradiction. It is the market telling you which part of the question it thinks is hard, and the next two sections are about how the rules produce it.

Live versions of these lines sit on the invasion market, the Kharg Island market and the Iran hub, all of which refresh every ten minutes.

What “invade” means in the Polymarket rules

The resolution text is one sentence and every word of it is load-bearing. The market resolves Yes if the United States commences a military offensive intended to establish control over any portion of Iran by 31 December 2026. For the purpose of the market, land controlled by Iran or the United States as of 4 November 2025 counts as that country’s sovereign territory.

Two clauses set the bar far lower than the word “invade” suggests to a reader. Commences means the offensive does not have to succeed, hold, or last; it has to start. Any portion means the target does not have to be strategic, or large, or the mainland. A single-island amphibious operation launched on 30 December and repelled on 31 December resolves this market Yes.

The Kharg contract is built the other way round. It resolves Yes only once Iran “no longer exercises primary governmental or military control” over the island and another state, occupying force or internationally backed authority has established control. Then it lists the exclusions, and the list is unusually explicit: temporary raids, isolated landings, special operations, bombardment, sabotage, naval presence offshore and temporary disruption of Iranian activity do not qualify on their own. Announcements and claims do not qualify without actual control. Contested or unclear control on the resolution date resolves No.

Ten points between commencing and holding

Read as a set, the four contracts price a chain of increasingly demanding events. The ratios below are not strict conditional probabilities — only one of the rows is close to a genuine subset of the invasion market — but they are the cleanest way to see what the board implies.

The Kharg row is the one worth arguing with, because it is not a subset of the invasion market and it still prices lower. The rules give control three routes: a US offensive, an internationally backed authority taking over, or a negotiated settlement, ceasefire term, surrender or transfer agreement — which would qualify once control was actually established, with no American offensive anywhere in the story. A contract with extra doors to Yes trading at a third of the contract that only has one door is the market saying it expects an attempt that does not finish, and that no diplomatic route ends with Iran handing over the terminal either.

Why Kharg Island is the contract that matters

Kharg is a small island in the northern Persian Gulf that handles up to 90% of Iran’s oil exports. On 13 March 2026 the US Air Force struck more than 90 Iranian military sites there, deliberately sparing the oil and gas infrastructure — the terminal was worth more intact as leverage than destroyed. Satellite imagery days later showed three tankers still moored. It remains under Iranian control today, which is why the near-dated rung is 0.25¢ with three days to run.

The island matters to the invasion question because it is the only place the two get welded together. In late March, reporting put the Trump administration on the verge of using ground troops to seize it — not as conquest but as coercion, a lever to force Tehran to reopen the Strait of Hormuz. CNN reported on 25 March that Iran was building up the island’s defences against exactly that. Four days later the Pentagon’s plans were described in detail: weeks rather than months, the 31st Marine Expeditionary Unit and the 82nd Airborne Division, raids into coastal areas near the strait, Kharg named explicitly, and stopping short of a full invasion.

That last clause is the whole trade. Stopping short of a full invasion is still an invasion under the Polymarket rules, and still not enough for the Kharg market.

The invade-Iran odds peaked at 67.5¢ on 30 March

The contract opened on 6 November 2025 at 13.50¢ and spent five weeks drifting down to a 6.00¢ trough on 9 December. The war began on 28 February 2026 with joint US and Israeli strikes. One month later, on 30 March, the market printed 67.50¢ — the day after the ground-plan reporting, and its all-time high. It has spent the six months since giving all of it back.

The shape of that path is the argument against reading 15.5¢ as complacency. This market has already shown it will reprice by fifty points in weeks when the reporting turns. It is not sitting at 15.5¢ because traders are not paying attention; it is sitting there because the specific escalation it was pricing in March was proposed, costed, and not carried out, twice — once before the April ceasefire and again after the July collapse. The 30-day move is +2.00 points, so it is not dead either.

Two ceasefires, thirty points apart

This is the part no page ranking for these queries covers, and it is the most informative thing on the board. Polymarket runs two separate ceasefire ladders — one on the Israel–Iran ceasefire, one on the US–Iran ceasefire — with rungs on the same four dates. They agree on this month and diverge violently after it.

Take the year-end pair at face value and the market is 65.5% confident the US–Iran ceasefire breaks before 31 December, while giving the Israel–Iran one a 64.5% chance of surviving. The war traders expect to come back is the American one, and they still price an invasion at 15.5¢. Those two facts only fit together one way: the expected resumption is more air and naval pressure and more blockade, not troops on Iranian soil. That is the same campaign the market has watched for seven months, and it is the reading that makes the whole board coherent.

What Kalshi lists instead of an invasion market

We swept 40 pages of Kalshi’s open events with nested markets and found five Iran-adjacent series. None of them is about an invasion, Kharg Island, or a declaration of war. There is no cross-venue check available on the biggest question in the conflict, which is the same situation as the Taiwan invasion market.

Kalshi’s substitute for a war market is an economic one, and it is the better-designed contract of the two venues: an eight-rung ladder on the seven-day moving average of transit calls through the Strait of Hormuz, resolving off IMF PortWatch rather than a consensus of credible reporting. It cannot be argued about, which is worth something on a question this contested.

Read the bid column and the curve is clean and steep: 29¢ that traffic gets above 30 calls a week at some point before the year is out, 2¢ that it gets above 100. The mids invert once, at Above 50, purely because that rung is quoted 16/26 — a ten-point spread on 6,568 contracts of open interest, the thinnest rung on the board. Use bids on this ladder, not mids.

2.65¢ that Congress ever votes on it

The quietest contract in the cluster is the one on constitutional process, and it has been falling all year: 10.50¢ when it opened on 13 January, 13.00¢ at its peak the next day, a 2.00¢ trough on 11 September, 2.65¢ now. Its rules are strict in a way that explains the level. Congress must pass a formal declaration of war under Article I, Section 8 and it must be signed into law. Authorisations for the use of military force, executive orders, presidential statements and military action itself — none of them qualify without the declaration.

So 2.65¢ against 15.5¢ is not the market doubting the war. Seven months of strikes, a naval blockade and two collapsed ceasefires have already happened without one. The contract is pricing the near-certainty that whatever comes next also happens without a vote, and the honest way to read it is as a measure of how little the formal instrument has to do with the fighting. Its live line is on the declare-war market.

What to watch before 31 December

The blockade ladder. Every reported Kharg plan had the same objective — reopening the strait — so the contract on Washington announcing an end to its blockade is the cleanest read on whether coercion is working. It is a well-ordered curve: 4.20¢ by 30 September, 16.50¢ by 15 October, 29.50¢ by 31 October, 45.50¢ by 30 November, 59.05¢ by 31 December, 74.00¢ by 31 March. The 30 September rung has fallen from 65.50¢ at the end of July, which is what a deadline slipping looks like in a price.

The midterms on 3 November. The conflict has pushed petrol prices up and is unpopular; the political cost of a ground operation is higher before the vote than after it. Watch whether the November and December ceasefire rungs diverge from the October one once that constraint lifts.

31 December itself. The invasion, both Kharg rungs, the declare-war contract and the regime-fall market all settle within minutes of each other, almost certainly at No. What gets relisted for 2027, and at what level, will be a better read on this conflict than anything trading today — a reopening price is a fresh opinion, not eleven months of drift.

Related on GeoOdds

• Will the US invade Iran before 2027 — live odds, chart, volume and the full resolution criteria
• Kharg Island no longer under Iranian control — the date ladder and the exclusions that decide it
• US announces end of the Iranian blockade — the six-rung curve on whether coercion is working
• Strait of Hormuz traffic returns to normal by 31 December — the shipping question, priced
• Iran regime change odds — two contracts, eight points apart, separated by one clause
• Strait of Hormuz reopening odds — why the ships lag the press release
• Iran odds hub — every active Iran market with 24-hour moves

Sources: the Polymarket Gamma and CLOB public APIs (events 73130, 267102, 160366, 72347, 711714, 1038648, 699735, 455875 and 830253) and the Kalshi public events and single-market APIs (KXIRANDEMOCRACY-27MAR01, KXUSIRANMEETING-27, KXTRUMPMOJTABA-26DEC, KXIRANIMPORTS-27FEB01 and KXHORMUZAVG-27JAN01), both read at 27 September 2026, 16:20 UTC; resolution-criteria quotations are from the market rules as published on Gamma and Kalshi. The 13 March strike, the 90%-of-exports figure and the post-strike satellite imagery are from the Kharg Island attack record; the Pentagon’s limited ground-operation planning, the named units and the Kharg seizure option are from Al Jazeera, 29 March 2026; the build-up of Iranian defences on the island from CNN, 25 March 2026; the operational case for and against a ground operation from Foreign Policy, 24 July 2026; and the Hormuz transit-call series that settles Kalshi’s ladder from IMF PortWatch. Prices captured 27 September 2026, 16:20 UTC and not updated since. Nothing here is investment advice — see our disclaimer.