What Actually Happens When A Contract Settles
Every beginner guide covers the first day of a contract's life: how to read the price, how to place the order. Almost nobody covers the last day, which is strange, because the last day is the whole point. So here is the ending, plainly: when a contract settles, every share of the winning side pays exactly $1, every share of the losing side pays exactly $0, and the question of which is which is answered by a data source that was named in the rules before anyone traded a single share.
That one sentence is most of the story, and we will unpack each piece of it. But the part worth staying for is the part that surprises people: what happens when the settlement number gets revised after the fact. There is a rule for that, it is close to universal on Kalshi, and almost nobody reads it until it costs them.
The Quick Answer
How do prediction markets settle? At expiration, the exchange grades the contract against the resolution source its terms named before trading opened: every winning share converts to $1 of cash in that trader's balance, and every losing share goes to $0. Delayed data means a late settlement against the same source, never a substitute. And once a contract has expired, a later revision to the underlying number does not reopen it, a finality rule we verified in the contract terms across Kalshi's jobs, inflation, crypto, Fed, weather, and sports markets. The three steps of settlement, that revision rule, and why finality matters most to sellers of long shots, are all below.
Settlement Is The Whole Point Of The Contract
If you have read our explainer on how prediction markets work, you know the machine: a yes/no question, shares trading between 1 and 99 cents, and a price that reads as a probability precisely because every share is headed for one of two destinations. Settlement is the moment the destination arrives. The lifecycle ends in three steps.
| Step | What happens |
|---|---|
| Expiration | The market stops trading; the question's answer is now a matter of record, or about to be |
| Determination | The exchange grades the contract against the resolution source named in the terms |
| Payout | Winning shares convert to $1 each, losing shares to $0, and cash lands in your balance |
The payout row is the one people ask about most, so let's be concrete about where the money goes. You do not need to claim anything or sell your winning shares. The position simply disappears from your account and cash appears in its place: $1 per winning contract, nothing per losing contract. From there it sits in your balance like any other cash, ready to trade or withdraw. The conversion is automatic but not always instant, and the gap between expiration and money-in-balance is its own subject; we covered the timing in why your Kalshi payout is not instant.
One more thing the table hides: settlement is the ending you can wait for, not the one you must. Any position can be closed early by selling at the current price. Settlement is simply what happens to whoever is still holding when the music stops.
The Source Is Named Before You Ever Trade
Step two is where the design earns its keep. Every contract is graded against a resolution source spelled out in its terms before trading opens. A jobs contract names the government report that decides it. A gas-price contract names the published national average it settles on. A temperature contract settles on the official figure from one named weather station, and the station is the contract: Houston's markets settle on Hobby Airport's instrument, not the bigger airport across town, and a weather app's "Houston" number is a different number than the one that pays.
Why front-load it? Because on an exchange, both sides of every trade are customers. For one to be paid $1, the other must be graded to $0 by a standard both accepted when they traded. Naming the source in advance converts "what happened?" from a debate into a lookup. It is also why the headline on a market can differ from what actually settles it; the terms, not the title, are the instrument. Our standing advice in this series: read them before you trade, every market, every time.
Delays, Revisions, And Ambiguity, Honestly
Naming a source does not stop reality from being messy, so let's cover the three messy cases without hand-waving.
Delayed data. If the source publishes late, the contract settles late, against that same source. Never against a substitute. A late settlement feels like a malfunction, but it is the system protecting the deal you actually made: late beats different, every time.
Ambiguous outcomes. Some questions can land in the gap between words and the world. When a determination is truly contested, the exchange has a review process under its rulebook, and because Kalshi is a CFTC-regulated exchange, its settlement conduct sits inside a federal framework rather than a customer-service policy. We wrote up what happens when a market is disputed separately; the short version is that the design front-loads the argument so that review stays rare.
Revised figures. This is the case people get wrong, and it deserves its own section.
A Revision After Expiration Changes Nothing
Official statistics get revised as a matter of routine. First prints of economic data are updated a month later, and even the public weather data these markets settle on occasionally gets corrected after quality control. So what happens to a contract that settled on the first number?
Nothing. That is the rule, and on Kalshi it is close to universal: once a contract has expired, a later correction to the underlying number does not reopen it. Most contract terms state that revisions "will not be accounted for in determining the Expiration Value." The high-temperature weather terms phrase it slightly differently, saying later corrections "will not be taken into account," but it is the same rule wearing different words. We checked the contract terms across jobs, inflation, crypto, Fed, weather, and sports markets, and the finality language appears throughout.
Walk through what that means in practice. Say a monthly jobs contract expires against a first print that lands inside one band of the ladder. The band pays $1, every other band pays $0, and the cash moves. Weeks later, the agency revises the figure to a number that would have landed in a different band. The settled contract stays settled. Nobody claws back a payout, and nobody gets a late one. The revision is next month's trading information, not last month's refund.
The rule can feel harsh the first time it goes against you. It is still the right design. Money has already moved between customers, and reopening settled contracts would mean no payout is ever final, which is worse for everyone than occasionally settling on a number history later adjusts. Finality is a feature. Price it in before you trade, because the terms already did.
A Worked Example: Why Finality Matters Most To Sellers
Here is the trade where settlement rules carry the most weight. A seller of an unlikely outcome collects a few cents of premium and puts most of a dollar at risk on every contract. Because the payout at settlement is always $1 or $0, that shape never changes: the win is small and frequent, the loss is large and rare. Roughly speaking, one loss erases the premiums from about 15 wins. That arithmetic, not the hit rate, is what makes position sizing the whole game, and it means a red day that wipes out a green stretch is the strategy behaving as designed rather than breaking.
Now connect it to this article. When most of a dollar rides on the outcome, you need the outcome to be a lookup, not a lawsuit. The pre-named source, the delay rule, and the no-revision rule are what let a seller price that risk at all. We say that as participants: Stokastic trades these markets and holds positions in them, and where this series describes a settled position, we were the seller. We do not publish performance figures on a permanent page like this one; the current picture of our open weather-trading experiment lives on the Kalshi weather markets hub, which is rebuilt through the day.
How This Compares To A Sportsbook Grading Your Bet
If you came to event contracts from sports betting, settlement is the piece with no real equivalent. A sportsbook grades your bet itself, under its own house rules, and it also sets the price you paid: a standard point spread posted at -110 on both sides carries a built-in hold of about 4.5% of the money wagered, and the fair price only appears once you strip that vig out. The book is your counterparty and your grader at once. On an exchange, your counterparty is another trader, the price is the crowd's probability, and the grader is a data source neither of you controls, named before either of you traded.
Neither model is charity, and the sharp habit is the same in both worlds: know exactly what standard your money is graded against before you put it down. In sports that means reading the market, then shopping the number across every major book on the live odds screen instead of accepting the first price. In event markets it means reading the terms, because as this whole piece has argued, the argument over your payout happened before you ever clicked buy.
FAQ: How Prediction Market Settlement Works
How do prediction markets settle? At expiration, the exchange checks the outcome against the resolution source named in the contract's terms. Winning shares pay $1 each, losing shares pay $0, and the cash is credited to your balance automatically.
Where does the money go when a contract settles? Into your exchange balance as cash. There is nothing to claim: the position is removed and $1 per winning contract appears in its place, ready to trade or withdraw.
What happens if the settlement data is delayed? The contract settles late, against the same named source. The exchange does not swap in a substitute source to settle on time.
What if the number is revised after settlement? The contract stays settled. Kalshi settlement is final at expiration: most terms say revisions "will not be accounted for in determining the Expiration Value," and a later correction does not reopen a settled market.
Disclosure and fine print. Stokastic trades Kalshi weather markets and holds positions in them; where a settled position is described in this series, we were the seller. We have no affiliate or commercial relationship with Kalshi, though we do carry sign-up offers for some other prediction-market and betting platforms. Kalshi contracts are CFTC-regulated event derivatives traded on a designated contract market, not sportsbook wagers, and a position can lose its full value. 18+, available where Kalshi operates; the risk of loss is real and, on the side we trade, individually large. This series is an open research log of a strategy we have not proven. Nothing here is trading advice, and nothing on this page is a pick or a recommendation.
