Government Shutdown Odds: October Was Easy. December Is The Cliff

Updated September 4, 2026 · 18 min read · by Eytan Shander

Will there be a government shutdown? For October 1, the market has all but moved on, and the record explains why. The House passed the Senate's stopgap 370-48 on Tuesday, September 1, and the President signed it into law the next day. The Kalshi contract on a shutdown at 10:00 a.m. ET on October 1 has closed its last three sessions at 1, 2 and 1 cents. Through mid-July it traded in the fifties. The August version of this page argued that 10 cents was too calm, because the two chambers had passed two different bills. The House then took the Senate's bill word for word, which is the cheapest possible way to make an argument like that wrong, and the market priced it before the gavel came down.

So this page has a new job. The stopgap, now law, funds the government through Friday, December 11, just over five weeks after the November 3 midterms and days before the Christmas recess, with zero of the 12 full-year spending bills through the Senate. The question moved 10 weeks down the calendar and changed shape on the way: it is now a count, on a Saturday-morning cliff, in front of a lame-duck Congress, where every House member and the senators on the ballot will already know whether they won or lost. Below is what the December board on Kalshi is quoting, what nine AI models made of the same question without seeing a price, and why the two are far closer this time than they were in August.

The Quick Answer

Not on October 1: the stopgap is law, funding runs through December 11, and Kalshi's October contract is bid 1 cent and asked 2, a one-to-two percent chance. The live question is December. Kalshi's board on how many shutdowns 2026 ends with puts a third one at roughly 32 to 37 percent, and nine AI models that scored the same question without seeing a price landed at 37 percent after a revision round. The full December board, every seat's number, and the four headlines that would move it are below.

Kalshi's Government Shutdown Odds Right Now

The October instrument is unchanged: a single binary contract on Kalshi, ticker KXGOVTSHUTDOWN-26OCT01, titled "Will the US government be shut down on Oct 1, 2026?" Every figure below was read from Kalshi's live market data at 10:39 a.m. ET on September 4, 2026. Quotes come from the order book; volume, open interest and the closing time come from the contract's own market record, and Kalshi's data feed reports volume and open interest as contract counts; at one dollar of face value per contract they convert one-for-one, so this page labels them face value throughout.

Market Read (As Of September 4, 2026)	Value
Yes Price (What Buyers Bid / What Sellers Ask)	1¢ / 2¢
Session Close, September 3	1¢
Session Close, September 2 (The Day The Bill Was Signed)	2¢
Volume, September 2 Session (Face Value)	$91,279
Lifetime Volume (Face Value)	$574,349
Open Interest (Face Value)	$258,195
Trading Closes	10:00 a.m. ET, October 1, 2026

A penny bid against two cents asked is as close to priced out as an open market gets, and this is a deep book. Roughly $574,300 of face value has traded over its life, more than double what had traded by August 24. About $258,200 is still open, open interest being the value of positions still held rather than closed out. The resting bid at a penny alone carries about $98,500 of face value, and the offers at 2 and 3 cents hold roughly $107,000 more. When a book this deep quotes one and two, the price is the crowd's probability, give or take spread and fees, and the market implies roughly one to two percent. What is worth studying is how it got from 55 cents to a penny, because the last two weeks of that slide happened before Congress cast a single vote.

Window (ET Sessions)	Where The Price Closed	What Happened In Washington
Mid-July	Fifties; 55¢ on July 14	Nothing resolved; a live coin flip
July 21	50¢, starting to drift	The House passed its own stopgap 220-205, funding through December 4
July 29 To 31	Volatile: 39¢, down through a 26¢ low, back to 37¢	Heavy two-way trading as the Senate deal took shape
August 2 To 3	17¢, then 18¢	The Senate's Collins-Murray stopgap text surfaced, funding through December 11; the White House budget office backed it on the 3rd
August 5	Back up to 33¢	A wobble; the book gave back most of the drop, then lost it again
August 7 To 8	22¢, then 17¢	The Senate passed its stopgap 90-6 on the 8th
August 9 To 21	A slow drift from 18¢ to 12¢	Both chambers in recess
August 22 To 27	8¢ on the 22nd, the low to that point, then back to 10¢ on the 23rd before settling 8¢ to 9¢	Recess; the busiest weekend the book had seen to that point
August 28 To 30	6¢, 6¢, 5¢	Recess ending; the House's return was days away
August 31	4¢ on about $55,800 of volume, the heaviest session to that point	The House came back; open interest jumped from about $207,000 to about $243,000
September 1	1¢	The House passed the Senate stopgap unchanged, 370-48
September 2	2¢ on about $91,300 of volume, the heaviest session of the last 30 days	The President signed the stopgap into law
September 3	1¢	Kalshi settled its full-funding ladder Yes on the signature

Two rows deserve a second look. The first is August 31. The House had not voted yet; the vote came the next day. Yet the busiest session to that point pushed the price from 5 cents to 4 and added roughly $36,000 of open interest in a single day. That is not a market reacting to a vote. It is a market pricing an expected one: by Monday evening the result of Tuesday's House vote was being treated as done, and the actual 370-48 tally moved the contract from 4 cents to 1. The second is September 2, the signing day, which traded about $91,300 of face value, the heaviest session of the last 30 days, and closed a cent higher than the day before, at 2. With the bid already at a penny the contract had nowhere lower to go, so the day's volume was the book changing hands rather than doubt about the law: Yes holders taking 1 and 2 cents to get out, and sellers writing the tail for them. Open interest tells the same story from the other side. It more than doubled between August 23 and September 1, from about $119,700 to about $254,500, while the price fell from 10 cents to a penny; across the three sessions since, through the signing and the settlement of the sibling ladder, it moved less than $4,000, to about $258,200. New money took both sides all the way down, most of it between 10 and 4 cents before the vote; what is left at a penny is a book that has stopped growing. Whoever is still writing Yes at 1 and 2 cents is collecting pennies, and somebody else is paying for the tail; the arithmetic of that trade is at the bottom of the page.

What Congress Actually Did, And What It Did Not

The mechanism the August version of this page leaned on was that the House and Senate had passed different bills: the House's ran to December 4, the Senate's to December 11 with the grants-rule delay attached, and nobody had reconciled them. The reconciliation turned out to take one vote. As NBC News and Breaking Defense reported, the House passed the Senate-passed text of H.R. 6500 with no changes, 370 to 48, sending it straight to the President. The bill keeps agencies at current funding levels, with a limited set of exceptions and program extensions, through December 11 or until full-year appropriations pass, whichever comes first. It also carries the Senate's delay of the Office of Management and Budget's federal-grants rule, but only until December 11, per the National Association of Counties' summary, which is a detail the December section below needs.

The signature followed within a day. The White House budget office had issued a Statement of Administration Policy on August 3 saying the Administration supports the Senate text and that "if this bill were presented to the President in its current form, his senior advisors would recommend that he sign it into law." On Wednesday, September 2, the White House announced that the President had signed H.R. 6500, the Continuing Appropriations and Extensions Act, 2027, into law. The December 11 deadline is no longer a bill; it is the statute.

Kalshi's other appropriations board confirmed the read, then closed. Its ladder on when a bill covering all 12 appropriations accounts becomes law, one contract per deadline from "before September 11" out to "before December 1," had repriced the September 11 rung from 85 cents to 99 in the session after the House vote; those contracts pay on a temporary extension as much as on a full-year bill, so on the afternoon of September 3 the exchange settled every rung on the ladder Yes on the signature. That leaves the October 1 contract as the only open market in Kalshi's shutdown series, and it says the same thing the settled ladder does.

That leaves the August panel's number to account for. Eight models, denied the price, put the October 1 shutdown at 27.5 percent when the market quoted 9 cents bid, 10 asked; the market has been closer so far. In fairness to the panel, its own watch table said what would prove it wrong: the House scheduling and passing the Senate's stopgap text unchanged was its top row, with a landing zone of 8 to 10 percent. That event happened, and the market went past 8 cents to 1. The verdict grades on October 2 when the contract settles, and it goes on the public scoreboard like every other call, wrong direction and all. What the August round got right is the part that matters now: the risk was never October. It was the deadline the stopgap creates.

The Cliff Moved To December 11

Fourteen. That is the number of days both chambers are scheduled to sit together between the November 3 midterm elections and the December 11 deadline, and it prices this board more than anything Congress says about its intentions. A continuing resolution, the CR in every Washington headline, funds the government temporarily and postpones the full-year fight; this one postpones it into those 14 days. Per the 2026 House and Senate calendars, the shared days are November 9 to 10, 17 to 20, and 30, then December 1 to 3 and 8 to 11, with the Senate adding a few solo days. Both chambers are also scheduled together December 14 to 17, which is after the deadline and matters only if a lapse spills past it; the House leaves for the holidays on December 17 and the Senate on December 18. Those 14 days are the lame-duck weeks, when members who have already lost or retired still vote, and the Senate walks into them without a single one of the 12 fiscal 2027 bills passed.

What has to fit into those days is the part that separates December from October. On October 1 Congress only had to extend existing funding levels, which is why the Senate produced a 90-6 vote and the House a 370-48 one. By December 11 it has to do one of two harder things. The first is finish the actual fiscal 2027 bills: per the Committee for a Responsible Federal Budget's tracker, updated September 3, the House committee has approved all 12, the House floor has passed 3, and the Senate committee has approved none, because it has not agreed on overall spending levels. Zero of 12 have passed the Senate; zero have become law. House Appropriations chair Tom Cole and ranking member Rosa DeLauro both said after the vote, per NBC News, that they want the bills done by year's end, and Cole added the caveat that matters most for the board below: "I do want to finish all the bills before the end of the year. Now, whether we can do that or not is another matter."

The second option is another stopgap, and it comes with a wrinkle that matters for the market below. If Congress enacts a second continuing resolution before any lapse, one that runs into January or February 2027, calendar 2026 ends with the two shutdowns it already has. The cheap escape is not just cheap; on Kalshi's December board it is the settling condition. The board noticed before Congress voted: the contract that pays on exactly two shutdowns this year closed the August 26 session at 55 cents and the August 29 session at 65, then reached 69 on the day of the House vote, on 34 contracts, about $34 of face value, that day. It printed 68 on September 3, on $21 of face value. December was repriced on almost no money, which is the reason the panel's number below is worth reading against it.

What The December Board Is Quoting

As of this read, the only open contract in Kalshi's shutdown series is the October 1 one; there is no contract on the government's status on the morning of December 12. The December read lives one board over, in the event titled "How many government shutdowns in 2026?" (series KXNUMSHUTDOWNS, event KXNUMSHUTDOWNS-27JAN01), one contract per count. Per the market rules, each contract pays on "the number of distinct government shutdowns (as defined in the GOVTSHUTLENGTH contract, with shutdown status checked at 10:00 AM ET each day)" in 2026, so a lapse counts once no matter how long it runs, and it counts only if it is in effect at that hour on some day this calendar year. Two have happened by that calendar-year count, which excludes the 43-day fiscal 2026 lapse that ended on November 12, 2025: January 31 to February 3, and the 76-day Department of Homeland Security shutdown that began February 14 and ended for most of the department on April 30, with ICE and Border Patrol appropriations left out of that bill and supplied instead by a reconciliation bill signed June 10. Kalshi counts that as one shutdown: the board lists no rung below "exactly two." The only way the count reaches three is a lapse in effect on some morning from December 12 through December 31. Prices read at the same 10:39 a.m. ET stamp as the October table:

Shutdowns In 2026	Yes Price (Bid / Ask)	Lifetime Volume (Face Value)	Open Interest (Face Value)
Exactly 2 (No December Lapse)	63¢ / 68¢	$33,252	$5,419
Exactly 3 (One December Lapse)	23¢ / 29¢	$21,823	$4,603
Exactly 4	3.4¢ / 6.5¢	$1,841	$1,557
Exactly 5	2.8¢ / 5.8¢	$749	$473

Read the first row in reverse and you have the market's December number: with "exactly two" at 63 to 68 cents, the board puts a third shutdown before New Year's at roughly 32 to 37 percent, using the inverse of that one quote rather than the sum of the others. Two caveats belong next to it. This is a much thinner book than the October contract, about $57,700 across all four rungs against $574,300 on one, and the "exactly three" contract traded about $2,400 on September 2 and $21 on September 3, so a single motivated trader can move it several cents. And the contracts are quoted a little rich taken together: the four asks sum to about $1.09, which is what a wide gap between bid and ask on a thin board looks like. Still, the shape is unmistakable. The market that just priced October to a penny puts December at one in three, and it has been drifting that way since late August.

The Verdict: Where Nine Blind Models Landed On December

We put the December count to nine AI models under the same protocol as August: none was shown a price, none was told what the others thought, and each received an identical briefing (the contract's settlement language re-fetched from the exchange, the legislative record through the September 1 vote, with the signature flagged as pending because it had not yet been reported, which does not move the December question since every seat treated the signature as the base case, the post-election session calendar, the fiscal 2027 bill count, and the full record of every shutdown since October 2025, each with its cause). Each returned a probability distribution across the four rungs and defended it. A second round followed, in which each model read the other eight analyses with the names stripped and could revise only with a stated reason. This time the seat that opened above 50 percent ended at 38, and the seat that started at 26 ended at 33. The panel ran on September 2; the seat numbers below are frozen from that run, and the market column is marked to the September 4 read.

The Board	Value
Market: A Third Shutdown In 2026 (From "Exactly 2" At 63¢ / 68¢)	About 32% to 37%
Panel Median After Revision	37%
Gap	Zero to about 5 points; the panel sits at the top of the market's range
Panel Range After Revision	33% to 40%
Market On "Exactly 3" (Bid / Ask)	23¢ / 29¢
Panel Median On "Exactly 3" After Revision	32%

Model	Round 1: Third Shutdown Or More	Final: Third Shutdown Or More	Final: Exactly 3
Claude Fable	38%	40%	34%
Claude Opus	36%	37%	32%
Claude Sonnet	53%	38%	32%
GPT	32%	37%	32%
Gemini	26%	33%	30%
Grok	35%	38%	32%
Kimi	38%	36%	31%
GLM	41%	40%	35%
DeepSeek	38%	37%	32%
Blend (Seat Median)	38%	37%	32%

Every seat on this panel is graded against real market settlements — records to date: Claude Fable 87% on 1,549 graded calls · Claude Opus 87% on 1,621 graded calls · Claude Sonnet 85% on 1,597 graded calls · GPT 85% on 7,942 graded calls · Gemini 86% on 6,624 graded calls · Grok 84% on 5,579 graded calls · Kimi 84% on 3,119 graded calls · GLM 82% on 2,941 graded calls · DeepSeek 81% on 2,985 graded calls. Recomputed daily; the full scoreboard is public.

These are model estimates, not predictions of fact and not financial advice. Every number in this piece gets graded in public once the market settles, and you can read the full graded scoreboard to see how each seat has actually done, including the October call above when it grades.

The headline is the agreement. In August the panel sat 18 points above the market; today the panel's 37 percent sits at the top of the range implied by "exactly two" at 63 to 68 cents, roughly 32 to 37 percent, and the whole disagreement fits inside a five-cent spread on a thin board. The models did not get cautious. Every seat built its number around the same asymmetry the briefing flagged: a second stopgap into 2027 settles "exactly two," and a Congress that cannot pass an omnibus, one bill carrying all 12 spending bills at once, in 14 session days has a cheap, familiar, already-demonstrated way out. GPT put the modal path plainly:

"The dominant path is another short CR into January or February 2027, because full-year appropriations are not close and that punt is procedurally easier than forcing an omnibus in the lame-duck window."

— GPT, 32% to 37%

What keeps the number near one in three rather than one in five is a detail that is easy to miss if you only look at the headline price. Claude Opus found it in the calendar, and it was the single most persuasive argument of the round: the deadline is a Friday, which makes the cliff a Saturday.

"The cliff is a Saturday, so a purely procedural overnight failure to clear the Senate on Dec 11 still books a countable lapse at 10 AM Dec 12 — the count doesn't care whether anyone missed a paycheck."

— Claude Opus, 36% to 37%

Four other seats cited that point by letter when they revised. It is why the panel's "exactly three" number, 32 percent, sits above the market's 23 to 29 cents while its "exactly two" number sits only a few cents under the market's ask: the models put more on a single short lapse than the board does, and less on two or more lapses inside the same three weeks, where the board asks about 12 cents across the two rungs against the panel's 5.

Put the gaps in trading terms and most of them disappear. The panel's 32 percent on "exactly three" against a 29-cent ask is three points of edge on a rung with a six-cent spread, a $21,800 lifetime book and $21 of trading in the last day; a buyer pays most of the disagreement just to get in, and pays it again to get out. The gap on "exactly two," 68 cents asked against a panel median of 63, is not holdable either, for a subtler reason. The seats think the calm rung is rich at 68, which is another way of saying they see a little more December risk than the ask implies. But the only way to trade that view is to sell, and selling means hitting the 63 bid, which is the panel's number to the cent. Rich on one side, fair on the other, and no edge on either. The honest summary of this board is that the market and the models disagree by less than the spread.

The biggest move belonged to Claude Sonnet, which opened at 53 percent for a third shutdown or more, the only seat above 50, on the argument that December is the deadline where the ICE and Border Patrol accountability fight, which the stopgaps have extended around rather than settled, has to be relitigated in a real fiscal 2027 DHS bill. Reading the others, it cut to 38. Its revision note adopts the new distribution outright and rejects the round's strongest opposing case, which, with the names stripped, was its own first-round argument:

"It overweights Congress's stated intent ('finish the bills') relative to its revealed capacity (0/12 Senate bills, tight lame-duck calendar) and its behavioral track record of choosing the plain-CR escape whenever schedule pressure is severe."

— Claude Sonnet, 53% to 38%

Gemini's numbers traveled the other way, from 26 to 33, with its second-round reasoning leaning on the unresolved ICE dispute and the OMB grants-rule expiry, which lands on the same December 11 date the stopgap does; the delay the Senate attached runs out exactly when the money does, so the fight neither side has resolved returns with the deadline rather than disappearing. Claude Fable, tied for the top of the board at 40, put the most weight on a channel most seats logged as a trigger rather than a base case: the President. The stopgap, by Breaking Defense's account, "does not include most of the anomalies," meaning special funding exemptions, that the White House requested, and a president who signed a clean bill in September that left out most of what he asked for can decline a clean bill in December. That is the December 2018 pattern, and Fable was not the only seat to name the year.

The cleanest statement of the case against a December lapse came from DeepSeek, which barely moved, 38 to 37, but pressed the election point hardest: the lame-duck session removes the campaign incentive that made hostage-taking pay in January and February, so the same members have less reason, not more, to force a lapse. GLM had logged the same fact cutting both ways in its first round, that the election lowers the cost of a lapse and removes the leverage that made one pay, and the round ended with the two halves of that argument at 37 and 40, the middle of the board and the top of it.

What Would Change The Panel's Mind

Every seat was asked to name the news that would most move its estimate, and the answers cluster tightly enough to read the run to December 11 by.

If This Happens	Direction	What The Panel Said
Leadership Announces A Second Stopgap Into January Or February Before Thanksgiving	Sharply down	The seats that named a number put "exactly two" at roughly 80 to 85 percent on that news
Either Party Publicly Conditions December Funding On ICE, Border Patrol Or Grants-Rule Language	Sharply up	The shape that produced the 76-day DHS shutdown; eight of nine seats named a version of it, none with a landing number
The President Signals He Will Not Sign A Clean December Stopgap Without Anomalies	Up	The December 2018 mechanism; Opus was the one seat to name a number, "exactly three" above 50 on a formal veto threat
A Stopgap That Moves The Cliff Only To December 18 Or 19	Up, and it fattens "exactly four"	A cliff still inside the counting window with Christmas on the other side; no seat put a number on it

Only the first row comes with a gradeable landing zone; the August panel did better on that score, and this one gets marked down for it when the board settles. The first row is also the cheapest signal and, as in August, it arrives well before the deadline. If the appropriators come back on November 9 talking about a January date, the December count is close to settled at two; if the first two post-election weeks pass with Cole and DeLauro still saying "finish the bills" and no stopgap text, the panel's number is the better read. When the "exactly three" rung gaps on a headline in between, check it against this table before calling it a trend; a thin board moves five cents on a rumor. (If your interest in prices runs toward games rather than Congress, our analysts publish free expert picks today on the sports side of the house.)

Date	What Happens
October 1, 2026, 10:00 A.m. ET	The October contract closes and resolves on the government's status at that hour; expected settlement October 2
November 3, 2026	Midterm elections
November 9, 2026	Congress returns from the election recess
December 11, 2026	Stopgap funding and the OMB grants-rule delay both expire at the end of the day
December 12, 2026, 10:00 A.m. ET	The first morning a lapse would count toward the 2026 total
December 17, 2026	Last scheduled House session day before the holiday recess; the Senate is scheduled through December 18
January 1, 2027	The 2026 count contracts close; any lapse beginning on or after this date counts zero

This page is re-scored when the story moves; the prices above are as of the morning of September 4, 2026, and the panel numbers are from its September 2 run.

What The Contracts Pay On, And What A Penny Buys

Two contracts, two different questions. The October one is the narrower: per the market rules it resolves Yes only if the federal government is at least partially shut down due to a lapse of appropriations at 10:00 a.m. ET on October 1, 2026, and the market rules list a formal Office of Management and Budget directive to "execute plans for an orderly shutdown" in effect at that hour, or an Office of Personnel Management operating status showing that "due to a partial lapse in appropriations, Federal Government operations vary by agency," as the examples that settle Yes. A technical lapse where OMB directs agencies to continue standard operations settles No. The December board is a count, and the count has its own trap: "exactly two" pays on the calendar year ending without a third lapse, whether or not Congress finishes its work, which a stopgap into 2027 achieves without settling anything. A trader who buys "exactly two" at 68 cents is betting on Congress punting or finishing; a trader who buys "exactly three" at 29 is betting on a lapse inside a specific 20-day window. That precision is the norm on Kalshi's economic boards, and it is the reason the December read has to be assembled from a count rather than read off a single contract.

Make the October price concrete. Buying 100 Yes contracts at the 2-cent ask costs $2. If the government is shut down at that hour under the definition above, they settle at $1 each: $100 back, a $98 profit before Kalshi's trading fees. If not, the $2 is gone. Two cents is a two percent implied chance, and reading Kalshi prices as odds is a translation of that probability; the contract itself is a CFTC-regulated event contract, in the same regulated family as Kalshi's election markets, 18+ and available where Kalshi operates.

Open interest on the October contract more than doubled while the price fell to a penny, most of it opened before the vote. Whoever sells that Yes at 2 cents collects $2 per hundred contracts against $98 of risk: a single losing settlement erases the premiums from 49 winning ones. At a penny, from 99. The sellers have the stronger case; the bill passed both chambers by veto-proof margins and is now law. But "stronger case" is exactly the phrase that ruins sellers of cheap long shots when they size as though it meant "certain," and this contract has already shown it can move 10 cents in a session. The December board is where that arithmetic gets interesting, because at 23 to 29 cents nobody is selling a tail. They are pricing an argument, and the argument is the one the panel and the market now share.

More live boards from the same panel: 2026 Midterm Election Odds · 2028 Presidential Election Odds — the shutdown boards themselves last quoted 1¢ bid on the October 1 contract and 63¢ bid on "exactly two" for the year; prices fetched September 4. The midterm board belongs next to this one: the stopgap runs to December 11 precisely so that the funding fight happens after the election, and every seat on the panel said the election result is the first input that changes its December number.

Hottest Prediction Markets Right Now

• 2028 Democratic presidential nominee · $188M traded
• 2027 Pro Football Champion · $68M traded
• 2028 U.S. Presidential Election winner? · $61M traded
• 2028 Republican presidential nominee · $59M traded
• Pro Baseball Champion · $58M traded
• Bitcoin price at the end of 2026 · $32M 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.

More on this: England Vs Pakistan Test Odds: Is The Kalshi Draw Cheap? · Florida Governor Odds: Donalds Vs Jolly After The Primary · 2026 Governor Races: Every Kalshi Board We Track · GTA 6 PC Release Date: The Pattern Rockstar Never Breaks · Jets Vs Titans Odds Week 1: Spread, Total & Kalshi Prices

The Bottom Line

The August version of this page argued with the market and is losing; the market said October was a scheduling formality and Congress made it one, and the panel's 27.5 percent takes its grade in public on October 2. The September version cannot argue much, because the two sides of the page have converged on the same fact: the cheapest thing Congress can do in December settles the December board, and the only way it fails is on a Saturday morning nobody planned for.

That is a different kind of risk from the one this Congress has produced three times since last October. Those were fights. December is a deadline, in a room full of people who have just been fired or re-hired, with the money and a grants-rule fight expiring on the same Friday. The price says one in three. The models say the top of that range. Both are betting on the same Congress to take the same exit, and both have written down what it would look like if it does not.

Disclosure and fine print.

Stokastic trades on Kalshi and holds positions in its markets; this series is an open research log of an approach we have not proven, and we do not publish our own trading returns. The model records quoted above are a separate thing: those are the panel's public forecasting scorecard, graded against settled markets. The shutdown contracts on this page are CFTC-regulated event derivatives traded on a designated contract market, not sportsbook wagers, and a position can lose its full value. Selling an unlikely outcome collects a small premium and risks most of a dollar; at the October contract's current price one loss erases the premiums from about 49 wins, and at a penny from about 99. Position size determines how much of that asymmetry a holder is exposed to. We have no affiliate or commercial relationship with Kalshi. 18+, available where Kalshi operates; the risk of loss is real. Nothing here is trading advice, a political forecast, a pick, or a recommendation.

Hero illustration: OddsShopper, in the house collage style. United States Capitol photo by Beethoven, licensed CC BY-SA 4.0; photos cropped, toned, and composited.

Trade prediction markets on Polymarket too.

Most readers here already have Kalshi — Polymarket is the other major venue, and code OS4 adds a $50 trading bonus on a $10 deposit: Get the Polymarket bonus →

Affiliate disclosure: OddsShopper earns a commission on signups through this link. Terms as stated by Polymarket; 18+, availability varies by state.