Skip to content

GuideHow-to · reviewed Sep 5, 2026

What Settles a New York City Weather Contract Is a Thermometer, Not a Forecast

Short answer. What settles a weather contract is a reading taken after the day ends, not the forecast before it. The two exchanges here name different New York City stations for the same day.

New York Predictions staff · Editorial desk · Last reviewed Sep 5, 2026 · 5 sources

What settles a weather contract, in one paragraph

A weather contract settles on an observation: a number measured by an instrument at a named place, published by a named source after the period being asked about has finished. A forecast moves the price and never settles the contract. That distinction is written into the rules rather than implied by them, because the two things can disagree by several degrees on the same afternoon and the money follows only one of them. The question a New York reader has to answer before reading any of these markets is therefore not what the weather will do, but which record the contract has agreed in advance to believe.

How a weather contract settles step by step

1. Find the station, because New York City is not one thermometer. The two exchanges quoting the city name different places for the same question. Polymarket's daily temperature markets resolve on the reading recorded by NOAA at the LaGuardia Airport station on the Queens waterfront, and point at the hourly observation table the National Weather Service publishes for it. Kalshi resolves its New York City rainfall market on the total precipitation at Central Park, and its daily high on the maximum temperature recorded at New York City, listed in its rules under the identifier CLINYC and reported by The Weather Company. A station beside Flushing Bay and a station in Central Park do not record the same afternoon, and neither exchange is wrong about its own contract.

2. Read the shape, because a bracket and a threshold behave differently. Polymarket writes the day as a set of brackets and resolves to the range that contains the reading, so exactly one of them pays and the rest expire at nothing. Kalshi writes the same city's high as a threshold instead, asking whether the maximum was greater than a stated number of degrees, which is a single yes-or-no line rather than a row of competing cells. Reading a threshold as though it were a bracket, or a bracket as though it were a threshold, misstates what the number is claiming.

3. Note the precision the source publishes. Polymarket states that its resolution source measures temperatures to whole degrees Fahrenheit and that whole degrees are the level of precision used to resolve, so a reading is never split between two bands by a decimal. Kalshi warns in its own rules that preliminary Weather Company data can differ from the final reported value through rounding and conversion, and tells readers to use caution with a preliminary number.

4. Read what happens when the number is late, revised or missing, which is where these markets have the most machinery. Polymarket accepts revisions to a day's temperature until the first data point for the following date is published, falls back to the Weather Underground daily observations table if the NOAA figure is unavailable by 11:59 pm Eastern on the day after the observation, and resolves to the lowest bracket if no data exists by then at all. Kalshi may hold expiration until a corrected publication where it decides there was a material error, and notes that a status of Final on its data provider's page is a convenience rather than a warranty.

5. Keep the two clocks apart. The reading is fixed at the end of the observation period; the contract expires later. A New York City daily temperature market closes at the end of its day and resolves once the first following data point exists or by late the next evening, whichever comes first. The monthly rainfall market for September 2026 stops trading on October 1, 2026 and carries an expiration of October 7, 2026, so it stays on the board for a whole month and then waits nearly a week to settle.

6. Remember that this board is a city, not a state. These contracts are written on New York City observations and settle on city instruments. Nothing on them describes a snowfall in Buffalo, a Hudson Valley frost or an ice storm in the North Country, and no reading taken upstate enters the calculation.

7. Check the window the reading is taken over. Polymarket's rule reads the highest figure in the temperature column for all times on the observation day, so an overnight spike counts exactly as a mid-afternoon one does, and a day that peaks at one in the morning settles on that peak rather than on anything that happened in daylight. The window is the calendar day at the station, not the part of it a reader was awake for.

Where New York readers stand

Whether either exchange quoting these observations is reachable by a person in New York is a question of what has been filed and what has been ordered, and it is not settled by a weather station. This guide takes no view on it and reaches no conclusion about any venue. New York's record on that question, with the date attached to each part of it, is on the legal status page. This guide is not legal advice and not financial advice.

Common misreadings

The forecast settles it. It does not, and no forecast appears in any of these rules. A published outlook is information about what the instrument will read, and the contract pays on what it did read. A contract can settle against every forecast that preceded it without any of them having been badly made.

A reading is a reading, so the station does not matter. It is the single most consequential line in a New York weather contract. One exchange settles the city's temperature at LaGuardia and the other settles it at a station recorded as CLINYC, and a day that lands on a band boundary at one of them can land inside a different band at the other. Two honest markets on the same city and the same date can settle differently for that reason alone.

A revision after the fact is too late to matter. It is explicitly not too late. Both exchanges write a revision window into the rules, one by accepting corrections until the next day's first data point and the other by holding expiration until a corrected publication appears. A number that looked final on the evening of the observation is provisional until that window closes.

A missing reading voids the contract. It does not. Polymarket resolves to the lowest bracket where no data exists by the stated cutoff, and Kalshi resolves at the last fair price it determined. A gap in the record is handled by a rule, and the rule is not always the one a reader would guess.

A band and a threshold can be compared side by side. Not without arithmetic. A bracket asks whether the reading lands between two numbers and a threshold asks whether it clears one, so several adjacent brackets add up to a threshold while no threshold decomposes into a bracket on its own. Comparing a bracket price on one venue against a threshold price on the other is comparing two different questions.

The city board itself is at the New York City weather page, and the weather vertical is at the weather hub. The threshold rows on the rainfall market are the same shape as a club's season, which is set out in the guide to how win total ladders settle, and what any single number on a weather band claims is covered in the guide to reading a probability.

This guide is not legal advice and not financial advice. It explains what a weather contract settles against and names no forecast, no contract and no exchange to use.

Common questions

What settles a New York City temperature contract?

A measured reading, not a forecast. Polymarket resolves on the temperature recorded by NOAA at the LaGuardia Airport station, and Kalshi resolves on the maximum temperature recorded at New York City, listed in its rules under the identifier CLINYC and reported by The Weather Company.

Which weather station does a New York City contract use?

Whichever station the exchange named, and the two do not agree. Polymarket names the LaGuardia Airport station in Queens for daily temperature, while Kalshi names New York City as CLINYC for temperature and Central Park for the monthly rainfall total.

Can a weather contract be resolved on a forecast?

No. Every rule on these markets names an observation source and a published reading after the fact, and a forecast appears in none of them. An outlook can move the price of a bracket all day and has no part in deciding which bracket pays.

What happens if the official reading is revised?

The revision counts inside a stated window. Polymarket accepts corrections until the first data point for the following date is published, and Kalshi may hold expiration until a corrected publication appears where it finds a material error in the first one.

When does a New York City rainfall market resolve?

After the calendar month it covers has closed. The September 2026 rainfall market on Kalshi stops trading on October 1, 2026 and carries an expiration of October 7, 2026, settling on the total precipitation measured at Central Park for the month.

Sources 5 · as the guide cites them

  1. Polymarket, highest temperature in NYC on September 5, 2026 (resolution source and revision rules)gamma-api.polymarket.com · primary ·
  2. Polymarket, lowest temperature in NYC on September 5, 2026 (the same rules on the low row)gamma-api.polymarket.com · primary ·
  3. Kalshi public markets API, series KXHIGHNY (New York City daily high temperature)api.elections.kalshi.com · primary ·
  4. Kalshi public markets API, series KXRAINNYCM (New York City monthly rainfall)api.elections.kalshi.com · primary ·
  5. Prediction Genius API, weather/temperature/nyc leaf (the daily high and low rows carried on the city board)predictiongenius.io · contract ·