The Quick AnswerAs of September 10, 2026, at about 9:15 p.m. Eastern, Kalshi's first rung on the 2026 WTI high, a settle above $115, trades at 45.4 bid and 46.3 ask. Wednesday morning it asked 29.5. WTI settled at $102.48 on Thursday, up $6.43 and 6.7% in a session, the highest close since May 19, after the Houthis took a Red Sea port within 50 miles of Saudi Arabia's bypass route and the first rung did what it had refused to do through three escalations: it repriced with the oil. Eight AI models re-priced the whole ladder price-blind Thursday night and average 59% on that first rung, about 13 points above the ask. The full ten-rung board, the hour the crowd finally moved, and where the seats split are below.
For three editions this page has carried one fact at its center: escalation after escalation, the market for a 2026 WTI settle above $115 would not move. On Thursday it moved. The front month settled at $102.48 on the NYMEX, up $6.43 on the day, per the AP. That is a 6.7% session, and it traded near $103 in the evening. Brent settled at $107.63 after briefly trading above $108, NBC News reported. That is WTI's highest close since May 19, both benchmarks are at their highest since mid-May, and both are up more than 75% for the year. Kalshi's first rung started the day asking 30.1 cents, printed 36.9 in the hour before 9 a.m. on about 19,800 contracts, and asked 46.3 at 9:13 p.m. The market that had priced the tanker war as a September story for a week decided Thursday that it is a 2026 story.
The number that did it was not a barrel count. On Thursday Yemen's Houthis, allied with Iran, seized the Red Sea port city of Mokha and pushed down the coast toward the Hanish Islands, less than 50 miles from the Bab el-Mandeb strait, Reuters reported. Bab el-Mandeb is the exit for the Saudi crude that has been leaving through Yanbu on the Red Sea since Hormuz was choked off, the bypass this page called "the route the Houthis sit on" on Wednesday. Thursday they moved onto it. That is the fact this edition is built on: for six months the crowd priced every escalation as a spike inside a range that one chokepoint, Hormuz, set. A second chokepoint under direct threat is a different range, and the year board, which sat still for eight tankers and a Brent breakout, repriced in a day. Eight AI models handed the same record Thursday night, without a single Kalshi quote, moved further than the crowd did: their average on the first rung went from 40% on Wednesday morning to 59%, and every seat that answered now calls the barrels lost rather than delayed. The full board is below, after the record that moved it.
What Thursday Changed
The record since Wednesday morning runs in three lines, and the third is the one the market's day lines up with. The first is the air war, and the price ignored it. Iran's overnight strike on Wednesday on the Muwaffaq Salti air base in Jordan on the Muwaffaq Salti air base in Jordan did more than first reported: about eight US F-15s were lightly damaged and an A-10 lost a wing, two sources told ABC News on Thursday, and Trump's answer at 11:59 a.m. Thursday was "Nope... None whatsoever. No damage. No nothing." Israel's defense minister, Israel Katz, warned that Iran's energy infrastructure would be struck if Israel came under fire, and the Revolutionary Guard said it destroyed a US Saildrone unmanned vessel in the strait, per CBS News; no new strike on an Iranian tanker was announced Thursday, so the count stays at eight since September 5. None of it is timestamped before 11:53 a.m., and the first rung had already crossed 36 cents in the hour before 9 a.m.
The second line is the clock. On Wednesday afternoon Trump told reporters, "I think the war will end immediately after the election, because they can't hold out any longer," per CBS, and said he was not seeking a deal with Iran; on oil, per NBC, "Right after the election, oil prices are going to be tumbling downward." On Thursday he said it again in a Fox News interview, per Reuters, adding "maybe sooner... I don't know that they're gonna be able to hold out." The election is November 3, the 38th trading session from Friday, with 40 more after it before the board closes. The Iran-Oman corridor that spokesman Esmail Baghaei said on Monday was in its final stages and days away has not been signed, and no new date has been given. Iran's spokesman spent Thursday on a different subject, rejecting the US description of the Houthis as a proxy: Ansarallah "takes orders from no one and acts as nobody's proxy." Pakistan delivered a Saudi warning to Tehran to restrain them anyway, and an Iranian official answered, "Iran does not control the Houthis," per Reuters.
The third line is the Red Sea. Mokha fell on Thursday. Saudi Arabia, which had struck about 40 Houthi targets across four Yemeni regions in the 24 hours before, per the Houthi-run Al-Masirah via ABC, issued its fourth emergency alert in a day for Khamis Mushait on Thursday. The UN's Yemen envoy, Hans Grundberg, warned of "a new and more dangerous phase." The flow numbers behind it got worse in the same 24 hours. Reuters counted seven vessels through Hormuz on Wednesday against a ten-day average of ten. OilPrice.com, citing Rystad Energy, put crude flows through the strait below 2 million barrels a day since September 2, against 6 to 9 million in early August, and said no very large crude carrier has exited the strait since September 2. Bloomberg reported that Saudi output fell in August to its lowest since 1990. On Wednesday this page set Macquarie's 7 million barrels a day against Energy Secretary Chris Wright's 9 million and called the whole board a bet on which number was real. Thursday's record carries a third number, Rystad's crude-only count of under 2 million, and a threat to the exit route that was supposed to make the other two survivable. ING's note read the price history plainly: "The move reflects a market still pricing in persistent geopolitical risk, with Persian Gulf tensions showing no credible path to de-escalation." The cost showed up at home the same day: diesel at $5.98 a gallon nationally and the 10-year Treasury at 4.95%, the highest since 2023, per NBC.
Kalshi carries three boards on WTI's price, and Thursday's tape on each one is the market's reading of that record. The quotes below are the exchange's book on Thursday, September 10, at 9:08 p.m. Eastern for the daily and month boards and 9:13 p.m. for the year board, and the prior prints are from its own hourly candles.
- Friday's Settle (The Daily Contract). With the front month near $103 in the evening session, a Friday close above $102.49 was bid 47 and offered 48; above $99.99 at 81 to 82; above $101.99 at 52 to 53; above $104.99 at 20 to 21. Holding Thursday's level is a coin flip; giving back $2.50 is priced at about one in five.
- September's High (The Month Board). On Wednesday morning the $100.01-or-above rung bid 39 cents under a 48-cent ask, and this page called it the rung the tanker war had repriced. Thursday's settle locked it in. It sits at 99 bid, with every rung through $102 at 99, and the live question has moved up the board: a September settle above $102.50 bid 91, above $105 bid 68 under an 82 ask, above $110 bid 28 under a 76-cent ask, with about 3,200 contracts open.
- The 2026 High (The Year Board). The $115.01-or-above rung had the day this page has been waiting for since September 2, and it reads better as a table than a sentence. Eastern time, hourly candles, each row the hour ending at the time shown.
| When | $115+ rung | The price history |
|---|---|---|
| Wed Sept 9, 11 A.m. | 27.7 / 29.5¢ book | The quote this page carried Wednesday (the book, not a candle) |
| Wed Sept 9, 2 To 3 P.m. | 30.1¢ close | WTI settles $96.05, up $3.02; about 390 contracts |
| Wed Sept 9, 3 To 4 P.m. | 30.5¢ close | About 1,700 contracts; Kalshi's close for the day is 30.4¢, 28.9 bid and 30.3 ask |
| Wed Sept 9, 5 To 11 P.m. | 30.2 to 30.4¢ closes | Trump's "immediately after the election" had been on the wires since 2:58 p.m.; about 3,850 contracts and the closes never left a two-tenths band |
| Thu Sept 10, 7 To 8 A.m. | 30.0¢ close, 28.9 / 29.9 book | About 640 contracts |
| Thu Sept 10, 8 To 9 A.m. | 29.9 open, 36.9 close, 36.4 / 37.1 book | About 19,800 contracts, the heaviest hour on any rung in two days of candles |
| Thu Sept 10, 9 To 10 A.m. | 40.0¢ close | About 2,980 contracts |
| Thu Sept 10, 11 A.m. To Noon | 41.9¢ close | About 1,060 contracts |
| Thu Sept 10, Noon To 1 P.m. | 41.6 high, 28.5 low, 38.4 close | About 5,040 contracts; a 28.5 print crossed and the book was back at 37.8 / 38.3 by the hour's end |
| Thu Sept 10, 1 To 2 P.m. | 37.9¢ close | About 1,110 contracts |
| Thu Sept 10, 2 To 3 P.m. | 37.5¢ close | WTI settles $102.48 at 2:30; about 1,620 contracts |
| Thu Sept 10, 3 To 6 P.m. | 40.2, 41.1, 44.9¢ closes | About 7,900 contracts across the three hours |
| Thu Sept 10, 6 To 7 P.m. | 54.5 high, 51.7 close | The day's high print, on about 4,040 contracts |
| Thu Sept 10, 7 To 8 P.m. | 51.0¢ close | About 830 contracts |
| Thu Sept 10, 8 To 9 P.m. | 44.4 low, 45.2 close, 44.9 / 45.9 book | About 1,440 contracts; the book was 45.4 / 46.3 at 9:13 |
Thursday put about 47,300 contracts through that one rung by 9 p.m., against about 9,700 on all of Wednesday and about 11,900 on Tuesday, the night the five tankers burned. The ask is 16.8 cents above Wednesday's 11 a.m. quote and 15.9 above Kalshi's own previous close of 30.4; eight of the ten rungs last traded above their previous close tonight; $180 or above sits a tenth of a cent under its close, and $200 or above, 1.8 cents under, is the one rung whose ask fell as well. The rung's open interest grew by about 12,400 contracts to about 584,100. That is Thursday in one line, and it is the opposite of the last three editions: the first rung spiked with the news, kept most of the spike, and by last trade finished the evening about 15 cents above Wednesday's close. Wednesday's edition said the crowd was waiting for a Saudi barrel count before it would pay for the first rung. It did not get one. It got a second chokepoint and paid anyway.
The rest of the ladder moved with it, and this time the tail moved too. Against Wednesday's quotes, $120 or above went from a 24-cent ask to 34.5, on a bid that rose from 23.7 to 32.6; in the 6 to 7 p.m. hour it printed as high as 43.7 before settling back. The $125 rung went from 22 to 30.7 and had the strangest hour on the board: about 9,600 contracts crossed between 6 and 7 p.m. with a 35.0 high and a 20.3 low, a 15-cent range inside 60 minutes on a rung that had traded about 470 contracts all of Wednesday. Then $130 from 19.3 to 27.2, $135 from 13.8 to 19.7, $140 from 12 to 18.3, $150 from 9.9 to 13.5, and $160 from 7.9 to 12.3, the rung that absorbed 15,400 contracts at 8 cents on Tuesday and is up more than half since. Two rows deserve their own sentence. The $200 rung, the cheapest on the board, got cheaper on the day oil broke $100: about 13,800 contracts crossed there between 10 and 11 a.m. at 5.2 to 5.6 cents, its heaviest hour in two days of candles, and it asks 7.4 tonight against 7.5 on Wednesday. And the odd print this page has carried since Tuesday, $160 or above offered cheaper than the less likely $180 or above, is gone: $160 asks 12.3 and $180 asks 9.1, back in the right order. The crowd repriced the middle of the ladder and left the far tail alone, which is what a market does when it moves its range up by $10 and not by $50.
Put the three boards side by side and the day reads differently from every day before it. The daily board repriced the level, as it always does. The month board locked in the rung it had bid at 39 cents a day earlier and moved its question to $105 and $110. And the year board, for the first time in this page's record, repriced with them: the first rung went from a 30-cent ask to 46, the second from 24 to 34.5, and the far tail stayed where it was. On Wednesday this page wrote that when the level rises and the near board rises with it while the year-high odds stay flat, the crowd is telling you it expects a spike, not a new range. Thursday the year-high odds did not stay flat. The crowd changed its answer.
The 2026 Oil High Kalshi Is Measuring Against
Kalshi's year contract does not care where oil finishes December. It pays on the maximum of WTI's daily settlement prices, the end-of-day price ICE reports, over the whole window. One qualifying settle on any trading day decides it for good, and an intraday spike that fades by the close counts for nothing.
Consider March 9. WTI traded as high as $119.48 that day and settled at $94.77, per Yahoo Finance's daily history for the front-month contract. Had Kalshi's rung paid on the touch, that day's print would have cleared $115. It pays on the settle, so the question is still open. The distinction is the whole contract, and Thursday is a smaller version of it: WTI traded near $103 after the close, but the number that counts is $102.48.
The 2026 record the panel was handed runs like this. The war between the US and Israel on one side and Iran on the other began February 28, and in the first full week of March WTI rose 35% to $90.90, the largest weekly gain since the contract began trading in 1983, per the Wikipedia chronology of the 2026 oil market; March 6 alone was a 12.2% session, the biggest one-day gain in the settle series since at least 2022. On March 14 the US struck military targets on Kharg Island, Al Jazeera reported. The year's highest close came on April 7, at $112.95, with an intraday high of $117.63 the same day. A ceasefire announcement around April 8 sent WTI down more than 16% to a $94.41 close in a single session.
What the forecast sites ranking for this query leave out is that the year then failed at the first rung a second time. WTI settled back above $100 on May 12, ran to $105.42 on May 15 and $108.66 on May 18, and was back at $98.26 two sessions later. Then it went the other way: the front month settled at $68.55 on July 6, its lowest close since the war began, two days before the US resumed strikes on Iran following attacks on three commercial vessels in the strait, per Al Jazeera. From that low to Thursday's settle is a climb of about 49.5%; from the August 26 close of $82.23 it is 24.6%, and every session since the August 28 close has been an up session, eight in a row. All told, WTI has closed above $100 in 18 sessions this year, above $110 in three, and never at or above $115.01. Only ten sessions this year closed higher than Thursday did, all of them between March 30 and May 19, and every one of them was followed within days by a close under $100.
So the first rung on the ladder asks for a settle 1.8% above what this same war already produced in April, and 5.8% above what it produced again in May. From Thursday's settle, it asks for a rally of 12.2%. On Wednesday morning that third number was just under 20%. The whole board repriced on the difference between those two, and the whole disagreement below is about whether a third run at the rung ends like the first two.
History since 2000 gives the crowd two precedents. Only two calendar years produced a daily close above $115: 2008, which peaked at $145.29 on July 3 on a demand boom and a weak dollar with no physical supply loss, and 2022, which peaked at $123.70 on March 8, when Russia's invasion of Ukraine put real exported barrels in doubt. Measured another way, starting from any daily close since 2000, WTI went on to close at least 12.2% higher within the next 78 sessions in about 51.3% of windows. Restrict the sample to windows that began with 21-session realized volatility above 40% annualized and the frequency climbs to about 63.5%. Realized volatility over the last 21 sessions is about 37% now, up from the 32.6% Wednesday's edition of this page carried, and still below that line. Those are the numbers that moved most between editions, and they moved for a mechanical reason: a 12.2% rally shows up nearly twice as often in the record as a 20% one. The panel got both, the reasons behind the two precedents, a warning that the unconditional rate counts rallies that started from troughs while this one starts from the top of a run the card described as five sessions and 10%, and an instruction to say which precedent's cause is present now. (The settle series says the run is eight straight up sessions since the August 28 close; the card line was corrected after the seats answered, so where a seat says "five-session" below it is quoting the brief as the models received it.)
How The Kalshi Oil Ladder Pays
The rule text on the first rung reads: "If ICE reports that the maximum price of oil (as defined exclusively by the set of WTI front-month settle prices) is above $115 between Issuance and Dec 31, 2026, then the market resolves to Yes." Every rung above it is the same sentence with a bigger number, prices are in cents and read as the crowd's probability, and a maximum market can only trigger once.
Which rungs are real markets is the useful question. The $115 rung is a two-sided book with about 584,100 contracts open, the largest on the board; about 3.22 million contracts are open across the ten rungs on 8.0 million traded since the board opened, and about 116,000 of those traded in the 24 hours to 9 p.m. Thursday, 48,000 of them on the first rung alone. The two busiest rungs after the first, by contracts traded, are still $150 and $180, not $120. The top of the board is thinner and the spreads say so: $180 or above sits nine-tenths of a cent wide and $200 or above 1.7 cents wide, so anyone trading the tail is paying a round trip that eats a meaningful slice of the position. Until a rung pays, its price should grind lower as the window shrinks, then gap on any supply headline, which makes this one of the twitchier boards Kalshi runs; Thursday was the gap. Thursday's settle is in the books, and 78 trading sessions remain through December 31. If you are new to how these commodity ladders are built, our guide to Kalshi's commodity markets walks through the daily, monthly and yearly boards and how each settles, and the Strait of Hormuz reopening board is the market that prices the ship count behind all of this.
The Ladder: Kalshi Vs Eight AI Models
Each model received the fetched settle history through Thursday, the 2026 record including the May run, the verbatim settlement rules, the base rates above with the causes behind the 2008 and 2022 exceptions, and the dated news record through Thursday evening, including Mokha, the Rystad flow figure and Trump's election timeline, and no market prices. Each produced one distribution across all ten thresholds in a single pass, so every column is internally consistent: the probability of $130 or above can never exceed the probability of $120 or above. The AI blend is the equal-weight mean of the seats that answered. Kalshi quotes are as of September 10, 2026, at 9:13 p.m. Eastern, and were moving as this was written; the September 9 column is the quote on this page at 11 a.m. Wednesday.
| Threshold | Kalshi bid / ask (Sept 10, 9:13 p.m. ET) | Kalshi bid / ask (Sept 9, 11 a.m. ET) | AI blend (Sept 10) | AI blend (Sept 9) | ChatGPT (GPT-5.5) | Claude Fable | Claude Opus | Claude Sonnet | Gemini 3.1 Pro | GLM 5.2 | Kimi K3 | DeepSeek V4 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| $115+ | 45.4 / 46.3¢ | 27.7 / 29.5¢ | 59% | 40% | 58% | 68% | 45% | 58% | 62% | 60% | 58% | 63.5% |
| $120+ | 32.6 / 34.5¢ | 23.7 / 24¢ | 45% | 28% | 43% | 57% | 32% | 42% | 49% | 45% | 42% | 49.7% |
| $125+ | 28 / 30.7¢ | 20.1 / 22¢ | 34% | 20% | 31% | 47% | 23% | 30% | 36% | 34% | 30% | 40% |
| $130+ | 24.5 / 27.2¢ | 17.4 / 19.3¢ | 24.5% | 14.4% | 23% | 38% | 16.5% | 21% | 26% | 24% | 20% | 28.7% |
| $135+ | 18.3 / 19.7¢ | 13.2 / 13.8¢ | 17.5% | 10.2% | 17% | 30% | 11.5% | 14% | 18% | 17% | 14% | 20% |
| $140+ | 18 / 18.3¢ | 11.8 / 12¢ | 12.5% | 7.3% | 12% | 24% | 8% | 9% | 12% | 12% | 10% | 14% |
| $150+ | 13.4 / 13.5¢ | 9.3 / 9.9¢ | 8% | 4.3% | 7.5% | 15% | 4.5% | 5% | 8% | 7% | 7% | 10.2% |
| $160+ | 11.1 / 12.3¢ | 7.8 / 7.9¢ | 4.4% | 2.3% | 4.5% | 8% | 2.5% | 2.5% | 4% | 4% | 4.5% | 5% |
| $180+ | 8.2 / 9.1¢ | 7.4 / 8.6¢ | 1.7% | 0.9% | 1.8% | 3% | 1% | 0.8% | 1% | 2% | 2.5% | 1.5% |
| $200+ | 5.7 / 7.4¢ | 6.5 / 7.5¢ | 0.7% | 0.3% | 0.8% | 1% | 0.4% | 0.2% | 0.5% | 0.8% | 1.2% | 0.3% |
Every seat on this panel is graded against real market settlements — records to date: GPT 86% on 9,653 graded calls · Claude Fable 89% on 2,856 graded calls · Claude Opus 89% on 2,995 graded calls · Claude Sonnet 88% on 2,971 graded calls · Gemini 87% on 8,676 graded calls · GLM 84% on 3,969 graded calls · Kimi 87% on 4,322 graded calls · DeepSeek 84% on 4,090 graded calls. Recomputed daily; the full scoreboard is public.
More live boards from the same panel: gas prices in 2026, where the $5.00-or-above rung asks 33¢ tonight · a Fed emergency meeting before 2027 · the full graded scoreboard behind every seat above.
Read the first row against its own history before you read it against the crowd. On $115 or above, the eight seats averaged 34% on September 2 from $90.95, 35% on Tuesday night from $94.60, 40% on Wednesday morning from $95.95, and 59% on Thursday night from $102.48, each from the WTI print that edition carried. The market, over the same four editions, asked 32, 29.4, 29.5 and now 46.3. Across the week oil closed about half the distance to the payout, the panel moved 25 points toward it, and the crowd, after three editions of moving the other way, moved 17 cents in a day. The gap between the two, two points on September 2, six on Tuesday and ten on Wednesday, is about 13 tonight, with the panel on top for the fourth edition running. One more number belongs next to it: in late July the crowd paid 45 cents for this rung against a panel at 28. The crowd is back where it was in July. The panel is 31 points higher. The next row down is where the crowd's confidence used to live: $120 or above asks 34.5 cents on a 32.6 bid, nearly two cents wide tonight against three-tenths on Wednesday, and the panel puts it at 45%, so the crowd now takes 11.8 cents off between the first rung and the second while the models take off 14. Then the two sides swap, as they have every edition. On $160 or above, which asks 12.3 tonight, the panel averages 4.4% and no seat is above 8%. On $180 or above, the crowd asks 9.1 cents and the panel averages 1.7%. The market is still paying real money for a scenario the models cannot find, and the models are now asking for a first rung the market prices near a coin flip. One of those two shapes is wrong, and the graded record will show which.
Model estimates generated September 11, 2026, at 01:13 UTC (9:13 p.m. Eastern on September 10; the three Claude seats answered the same brief at about 9:40 p.m. Eastern), price-blind. Models are frequently wrong; the market price reflects real traders' money.
Hottest Prediction Markets Right Now
- 2028 Democratic presidential nominee · $214M traded
- 2027 Pro Football Champion · $86M traded
- Pro Baseball Champion · $69M traded
- 2028 U.S. Presidential Election winner? · $65M traded
- 2028 Republican presidential nominee · $64M traded
- Fed decision in Sep 2026? · $41M traded
Every market above links to our full AI model verdict; browse them all on the OddsShopper prediction markets hub, and see how every settled call actually scored on the full graded scoreboard.
Where The Panel Splits
Wednesday morning, seven of the eight seats called this week a physical loss of barrels. Thursday night it is eight of eight, and the holdout that flipped was the seat that had made the best argument for holding out. Every seat raised its number on the first rung, and the moves run from 10 points to 34. Claude Fable, which came DOWN to 34% on Wednesday because "this identical war twice stalled below $113 settles this year," doubled to 68% and leads the board; Gemini went from 45% to 62%, DeepSeek from 45% to 63.5%, GLM from 38% to 60%, and ChatGPT, Kimi and Claude Sonnet all landed at 58%, from 44%, 45% and 33%. Claude Opus, at 45%, is the low seat, up from 35%. The reason is the same in every seat's own words, and it is Thursday's third line. Fable: "the second chokepoint — Houthis 50 miles from Bab el-Mandeb threaten the Yanbu bypass, which is what capped May." Gemini: "Unlike the April and May peaks which were largely risk premiums, millions of barrels are now physically trapped." Sonnet, which held its number at 33% on Wednesday while naming the corridor as the thing that would cut it, now sees "a second chokepoint (Bab el-Mandeb) now live via the Houthi advance, removing Saudi's Red Sea bypass as a clean escape valve," and adds the clock: Trump "explicitly ties de-escalation to after Nov 3, leaving ~7 weeks of open-ended escalation risk before any relief."
Most of that move is arithmetic before it is judgment. The base rate handed to the panel did most of the work. A 12.2% rally inside 78 sessions has happened in 51.3% of windows since 2000 and 63.5% of high-volatility windows; Wednesday's edition, from $95.95, carried 28.2% and 43.4% for the 19.9% rally then required. Six of the eight seats landed between Thursday's two numbers, DeepSeek on the upper edge, with Fable above the band at 68% and Opus below it at 45%, and the average of 59% sits closer to the high-vol one. DeepSeek took the high-vol rate outright: its $115, $120, $130 and $150 rows, 63.5, 49.7, 28.7 and 10.2, are the high-vol base rates to the decimal. Kimi "blended toward the 63% high-vol rate for regime," then took about ten points off because the window starts at the top of a 24% run and the year has already failed twice at the first rung, added some back for the worsening chokepoint math, and checked the result against "a driftless-GBM touch estimate of ~57%." ChatGPT "haircut the unconditional 51% base rate upward modestly for live chokepoint risk, but not to the high-vol 63.5% because the contract has already failed near $113 and $109, and policy brakes are credible." GLM built the off-ramp into its number: "roughly 35–40% odds of a politically forced ceasefire before Dec 31 that would cap prices short of $115; absent that, the physical shortage and five-session momentum make $115+ likely." Fable went the other way and shaded the high-vol rate up to 68 "because the loss is now physical," and put a fatter tail under it because "Gulf spare capacity trapped behind Hormuz is not a brake — only SPR releases and demand destruction (diesel $5.98) are."
Opus is the seat to sit with, because on Wednesday it named its price. It was the one seat still calling the week a premium, and its argument was that "Hormuz flows are already halved yet WTI sits near $96, so $115 needs a new loss of Saudi bypass barrels, not more of the same." Thursday delivered the threat to exactly those barrels, and Opus moved ten points and flipped its reading: flows under 2 million barrels a day "and the Yanbu–Bab el-Mandeb bypass now under threat, can break the April high." Then it stopped at 45, below every other seat, below the 51% unconditional rate, and below its own "driftless barrier model at 37% vol, about 55% to reach $115," and gave four reasons: "Daily settles miss intraday spikes. The rally starts from the top of a five-session run. This war has failed at this level twice. And Washington has strong midterm incentives: SPR releases, pressure on Iran and Oman to sign the corridor, and Trump's post-election end date." That is the crowd's case, stated better than the crowd states it, and it is why the low seat on the panel sits within half a cent of the market's bid tonight. The other seven think the barrier model is closer to right than the failures are.
The seats agree about the top of the ladder more than the bottom, as they have every edition, and this time the top moved with the bottom. Asked for the highest settle they would expect in the worst one-in-twenty version of the rest of 2026, the eight answered between $148 and $165; Wednesday the range was $142 to $155. Seven of the eight raised their worst case, Opus to $152, Sonnet $153, Gemini $157, Fable $158, ChatGPT and DeepSeek $160, Kimi $165; GLM is the exception, and cut its from $155 to $148. Not one put that number above $165, which is still the whole explanation for the tail rows: a bad case that tops out near $160 makes $180 or above at 9.1 cents and $200 or above at 7.4 prices for a different world than the one any seat is modeling.
Skip past the first rung for a moment, because $130 is the row where the two sides met on Thursday. On Wednesday the crowd asked 19.3 and the panel said 14.4%, within five points for the first time this month. Tonight the crowd bids 24.5 and asks 27.2, and the panel's blend, 24.5% in the table and 24.65% unrounded, sits on the bid. Fable at 38% is above the ask, Opus at 16.5% is below the bid, and the other six sit between 20% and 29%. On the first rung the two sides are 13 points apart; on $130 they are on top of each other; on $160 they are 8 cents apart the other way. The market's shape, read across those three rows, says it sees Thursday as a thing that might add $15 to the year's high and has not decided whether it adds $30. The panel's shape says it is more sure about $115 and less sure about everything above it.
What Moves This Board Next
Three things would move every number on this page, and the seats named all of them, with different weight. The corridor comes first, and it cuts down. Six of the eight put a signed Iran-Oman transit deal, or the ceasefire it would stand in for, at the top of the list of what would most cut their numbers. Kimi put a figure on it: a signed, IMO-documented corridor with the US accepting it "collapses p1 to ~25%." Fable said it "could replay April's 16% one-day collapse." GLM's 35-to-40% odds of a "politically forced ceasefire before Dec 31" is already inside its 60%. Baghaei's days-away corridor is now three days old with no date, Washington's objection to Iranian-Omani fees is the friction on it, and Trump's own timeline runs the other way, to November 3, 38 sessions out. The Red Sea comes second, and it cuts up. Opus, the low seat, said the news that would most change its estimate is "a Houthi strike on, or closure of, Bab el-Mandeb or Yanbu," and Fable and Sonnet built their numbers on the threat to those barrels. Nobody has published a Saudi barrel count since Tuesday's fires, and the Kingdom's August output at a 1990 low is the closest thing the record has. ChatGPT named the reverse tell: "verified Hormuz crude flows returning above 6 million b/d for several days" would most lower its number, which puts Rystad's sub-2-million figure at the center of the bull case. The near boards are the third tell, and the fastest. Friday's settle above $102.49 is bid 47 under a 48 ask; a September settle above $105 is bid 68 under 82, and a print there would cut the first rung's remaining distance from 12.2% to 9.5%. Watch the $105 month rung the way this page watched the $100 rung on Wednesday. That one locked in inside a day.
The same shock runs through two other boards this panel prices. The gas price board marks a $5.00-or-above national average at 33 cents tonight, and is where a $102 crude settle lands next, and the Fed's rate path is where a 4.95% ten-year already has. Prices on this page refresh during the news wave; re-run the payout arithmetic at the price you are actually offered, because the 46.3-cent ask quoted here printed 54.5 and 44.4 inside the same three hours.
Zoom back out and the ladder reads simply. The year has already shown twice what this war does to WTI: a $112.95 settle in April, about $2 short of the first payout, and a $108.66 settle in May, each erased within days. For a week the crowd priced the third run the way the first two ended, and for a week this page recorded a first rung that would not move. Thursday it moved about 16 cents, on 47,000 contracts, because the war reached for the one route that had made the other two runs survivable. Eight models handed the same record moved further, to 59% on a rung the crowd now prices near a coin flip, and the tail is the one place they still disagree the old way: the crowd asks 12.3 cents for $160 and the models average 4.4%. The number this board needs next is still a Saudi barrel count, and it is now a count on both coasts. The crowd stopped waiting for it on Thursday. The year board's price says a third run at $115 is no longer the long shot, and 78 sessions will say whether that was the right day to decide.
Stay ahead of the markets.
Daily insights and expert picks on Kalshi, Polymarket, and what's moving markets.
Free forever. Unsubscribe anytime.
Hero illustration: OddsShopper, in the house collage style. Strait of Hormuz photo by MODIS Land Rapid Response Team, NASA GSFC, licensed Public domain; photos cropped, toned, and composited.
Prices on this page are Kalshi's book. If you also trade on Polymarket, code OS4 gets new users a $50 trading bonus on a $10 deposit — affiliate link; terms as stated by Polymarket; 18+, availability varies by state.
