The National Weather Service Data Kalshi Settles On
Every Kalshi temperature contract settles on National Weather Service data that anyone can read, free, at the same moment. Not a private feed, not a subscription product, not a number a market maker sees before you do. A United States government agency measures the temperature at a named station, publishes the official daily report, and that report decides which band pays $1 and which bands pay $0.
Almost nobody who writes about these markets covers where that number actually comes from. This article does. By the end you will know what a CLI report is, how to read one line by line, where to find it in about ten seconds, and why a settlement source like this makes weather unusual among prediction markets. One honest caveat waits at the end, because equal data is not the same thing as equal results.
The Quick Answer
The settlement data behind Kalshi's temperature markets is the National Weather Service's daily climate report, called the CLI, which publishes the official high and low for one named station and is free to everyone with no account and no paywall. That report decides which band settles at $1 and which settle at $0. Because the source is a public agency reading a public instrument, no participant in the market has earlier or better settlement data than you do. What a CLI report contains, how to decode its temperature block, and where the symmetry stops are all below.
The Scoreboard Is A Government Document
Start with what settlement actually is. A Kalshi temperature market splits a city's daily high or low into bands, and the whole ladder trades as a probability distribution. When the day is over, one official number decides everything: the high or low recorded at the single weather station named in that market's rules. The band containing that number settles at $1 per contract. Every other band settles at $0. These are event derivatives on a CFTC-regulated exchange, which is what the regulation actually means, and binary settlement needs an unambiguous source.
That source is the National Weather Service. Each market's rules point at the official daily climate report for a specific station, so the scoreboard for the entire market is a government text product. We have written before about why the station itself is a trap for new traders: Houston grades on Hobby Airport, not Bush Intercontinental, and a weather app's city reading is a different instrument entirely. This article is about the report those rules point to, because once you can read it, you are looking at the same page the contract grades on.
What A CLI Report Is
CLI is the National Weather Service's product code for the Climatological Report (Daily). Every local NWS forecast office issues one for each of its designated climate stations, which are mostly automated instruments at airports. The report is plain text, a few dozen lines long, and it is the official daily accounting of what that station measured: the maximum temperature and when it occurred, the minimum and when it occurred, precipitation, snowfall, wind, humidity, degree days, and how each figure compares to normal and to the records.
Two details matter for a trader. First, the report names its station in the header, which is how you confirm you are reading the thermometer the contract grades on rather than one across town. Second, more than one version of the report can exist for a day: many offices issue an intermediate report with numbers as of that hour, and the final report, issued after the day closes, covers the full calendar day. The market's rules say which report settles the contract, so the rules page and the report together are the entire settlement chain. No other feed is part of it.
The instrument behind the report matters too. The station's readings come from an automated sensor suite that reports around the clock, so the daily maximum in the CLI is not an estimate or a model output. It is the highest value that instrument actually recorded. A running maximum can only rise and a running minimum can only fall, which is why the high book and the low book resolve at opposite ends of the day.
How To Read One, Line By Line
Open a CLI report and the block you care about is labeled TEMPERATURE. It reads like a table with the day's value, the time it occurred, and the context columns. Here is the block decoded, using the shape of a summer afternoon at a settlement station:
| Line In The Report | A sample reading | What a trader learns from it |
|---|---|---|
| MAXIMUM | 97, at 3:47 PM | The settlement number for the high market. The band containing 97 pays $1; every other band pays $0. |
| Time Of Maximum | 3:47 PM | When the day was decided. A mid-afternoon peak is typical; the time column teaches you each city's rhythm. |
| NORMAL MAXIMUM | 91 | The climatological baseline, so 97 printed 6 degrees above normal. Context for how unusual the day was. |
| RECORD MAXIMUM | 103, with the year it was set | The historical ceiling. Useful for judging how far the tail bands really are from reality. |
| MINIMUM | 76, in the pre-dawn hours | The settlement number for the low market, which is a different game on a different clock. |
The actual report packs the value, time, record, normal, and departure into columns on each line; they are unpacked into rows here for clarity.
The single most useful row for a newcomer is the time of maximum. Most people picture a daily high as a fact about the whole day. The report shows it is a fact about one moment, usually mid-afternoon, and everything after that moment is irrelevant to the high book. Watch that line across a few weeks of reports and you will understand a city's daily rhythm better than most of the people trading it.
Below the temperature block, the same report carries the precipitation block, which is the official accounting that rain markets grade against, plus wind, humidity, and degree days. One document, one station, every settlement-relevant fact for the day.
Free Means Free
Here is the part that would sound strange to anyone from traditional finance: the settlement data is public and free. The National Weather Service publishes CLI reports on its own websites, including each forecast office's climate page at weather.gov, and through its public API. There is nothing to sign up for and nothing to pay, and a premium tier that gets the number faster simply does not exist. The document deciding whether a contract pays $1 or $0 costs you $0 to read. The agency exists to publish this data to everyone, it has no commercial stake in any market, and its reports are in the public domain.
Compare that to the data economics almost everywhere else. Sports betting runs on injury news, lineup leaks, and paid data feeds that move faster than public reporting. Equity markets sell co-location and direct feeds precisely because milliseconds of earlier data are worth money. In weather, the official scoreboard is a taxpayer-funded text file that publishes to the whole world at once. You, reading this, have exactly the same settlement source as the largest trader on the exchange.
No Insider Has Better Data Than You
That leads to the unusual part, and the reason this cluster keeps returning to weather as the cleanest teaching market. In most markets, information asymmetry is the game. An election market rewards whoever saw a private poll. A sports market rewards whoever heard about the injury first, which is a real difference between prediction markets and sports betting. Financial markets police insider information with an entire body of law because insiders exist.
Weather has no equivalent. There is no locker room, no campaign war room, no embargoed report to leak, and no way to bribe the atmosphere. The forecasts most traders lean on are themselves publicly funded and published to everyone. Nobody can preview tomorrow's thermometer reading, and once the readings start arriving, they arrive in public, in near real time, from the same instrument for every participant. Whether a band is trading at 7 cents or 70, every trader on both sides of it is pricing the same public report. A careful newcomer is not structurally behind on information here, and there are very few markets anywhere you can say that about. It is a real part of why prediction market prices track outcomes as well as they do: when the settlement source is public, the argument is only ever about interpretation.
The honest caveat: equal data is not equal skill. Two people can read the same CLI report and price a band differently, because one of them understands how much afternoon is left, how this city behaves in this season, and what a fair price for the tail is. Data symmetry removes one class of disadvantage, the structural one. It does not remove the work, and it does not make the market beatable by default.
What Symmetry Does Not Save You From
The work is mostly risk, and it deserves its own paragraph because it is the failure mode we care most about. A common style of trade in these markets is selling an unlikely band: collect a small premium, risk most of a dollar. Roughly speaking, one loss erases the premiums from about 23 wins. That arithmetic, not the hit rate, is what makes selling long shots such an asymmetric proposition, and it means position sizing matters more than being right. A red day that wipes out a green stretch is the shape of this style of trading, not a malfunction, and no amount of public data changes it.
We say that from experience rather than theory. Stokastic trades these markets and holds positions in them, and where a settled position is described in this series, we were the seller. You will not find our results on this page, deliberately: this page is permanent, and any figure printed here would freeze while the real log kept moving. The shape is what we can tell you. The log is short, it sits in the red so far, and it is far too small to confirm the edge or to rule it out. Current figures live on our Kalshi weather markets hub, which is rebuilt several times per day, and nothing in this cluster is a pick.
Read The Primary Source
Zoom out and this article is one idea. The number that settles a Kalshi temperature contract is National Weather Service data, published free to everyone, in a document you can now read: the CLI report, one station, one official high and low, with the time-of-maximum line quietly teaching you when each city's day is decided. Weather markets are unusual because their scoreboard is public infrastructure, so the entire contest is interpretation and discipline rather than access. If you are brand new and want the wider context first, start with how weather markets actually work.
The transferable skill is reading the primary source instead of someone's summary of it. That is the same reflex that makes line shopping work in sports betting: the same game priced at DraftKings and FanDuel only means something once you pull both numbers up side by side on an odds screen and read them as probabilities, and the fair number only appears once you de-vig the book's hold out of them, the way the real settlement number only appears once you find the station's own report. Different market, same discipline: go find the document the outcome actually grades on, and read that instead of the summary.
FAQ: National Weather Service Data And Kalshi Settlement
What is a CLI report? CLI is the product code for the Climatological Report (Daily), so an NWS CLI report is the plain-text product each local National Weather Service office issues for its designated climate stations. It carries the official daily maximum and minimum temperature with the times they occurred, plus precipitation, wind, humidity, and comparisons to normals and records.
Is the settlement data really free? Yes. CLI reports are public-domain government data, published on weather.gov and through the NWS public API with no account and no paywall. Every market participant reads the same report at the same time.
When does the CLI report come out? Issuance varies by forecast office: many publish an intermediate version during the day with figures as of that hour, and the final version after the day closes covers the full calendar day. Each market's rules state which report settles the contract, so check the rules rather than assuming a schedule.
Does Kalshi settle on my weather app's temperature? No. Contracts grade on the official reading from one named station in the market's rules. An app's city temperature is a blended or nearest-sensor number and can differ enough to flip a band.
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. Kalshi weather 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 yet proven. Nothing here is trading advice, and nothing on this page is a pick or a recommendation.
