BuildNot yet confirmed elsewhere1 publisher2 min readPublished
A Kalshi series titled SATX settles on Louisville's CLISDF station
A dev.to developer reports that Kalshi's API titles its Louisville temperature series 'HIGHEST Temperature SATX' while the event and rules name Louisville. A join on titles files Louisville contracts under San Antonio; the station code in the rules is what pays out.
The Engineer · Build desk
Key figures: Louisville's market resolves Yes if the maximum recorded at Louisville (CLISDF) is less than 82 degrees Fahrenheit, according to The Weather Company. In the developer's Chicago row, the model forecast a high of 86.2, and Midway had already recorded a high of 88.
| Figure | Value | Note | Claim |
|---|---|---|---|
| Louisville (CLISDF) Yes if high is less than | 82 °F | 5 | |
| Chicago forecast high | 86.2 degrees | 13 | |
| Midway recorded high | 88 degrees | 14 |
What happened
- The author's call to the public series endpoint, filtered to Climate and Weather, returned 422 entries on 12 October 2026, and the post prints two of them.
- San Antonio has its own series, KXHIGHTSATX, titled 'San Antonio Daily Maximum Temperature', so the SATX label appears on both a correct series and a Louisville one.
- The sibling low-temperature series, KXLOWTSDF, is titled 'Lowest Temperature SDX', which is not a code for anywhere.
- The Louisville event's rules cite station CLISDF and resolve against The Weather Company's recorded maximum, with a threshold of 82 degrees Fahrenheit.
- Newark and Trenton are separate series, KXHIGHTEWR and KXHIGHTTTN, so New York weather is three separate ladders.
Why it matters
- constraint Any pipeline that joins markets to cities by title text will misfile Louisville contracts under San Antonio. The city key has to come from the settlement station, one lookup per market.
- decision Builders need a station table, a parser that returns nothing on a miss, and the title demoted to a display label. A fuzzy match on titles would hide the mismatches this one exposes.
- exposure The one reliable input is free text. If Kalshi rewords its rules or adds a second parenthesised code, a regex over prose breaks or misreads, and the post shows no structured station field to fall back on.
- capability The CLI-to-K swap gives a free reading from the settling station. A builder can check a model-versus-market gap against the recorded high before calling the gap an edge.
The author's fix is a one-line parser that ignores titles. It reads the rules text for a parenthesised code and looks the city up in a station table [8]. The author wrote, "The title is a fallback label and nothing else." [7]
The pattern accepts strings of five to seven characters [16]. CLISDF fits. So would any other parenthesised code of that shape elsewhere in the rules. When nothing matches, the function returns null [8]. In our view that is the right default: a market with no station gets no city.
The evidence for the wider claim is thin. The post prints two of the 422 series entries and documents one mismatched title and one garbled sibling, all from calls made on 12 October 2026 [1]. It does not say how many titles disagree with their rules.
The city matters because of how these markets settle. Each one pays out against one climate report from one point on the map [10]. The author measured what a wrong guess costs. Both points came from Open-Meteo, two coordinate pairs per city, daily highs for the next three days, pulled on one afternoon [18]. Phoenix and Washington stayed within half a degree. The post says the gap is coastal and lake-effect geography and is biggest where the volume is [19]. Kalshi's brackets are two degrees wide, and the author calls a 4.2-degree error two whole brackets [20].
That figure comes from one provider, three days and one afternoon. It transfers only if your fallback coordinates miss the settling station by as much as the author's alternate pair did.
Observations have a shortcut. Swap the CLI prefix on the climate-report code for K and you have the NWS station id: CLIMDW becomes KMDW, and the NWS observations endpoint for that id returns what the station has recorded [11]. Settlement itself is per The Weather Company, so the author treats NWS data as a sanity check and not the official number [12].
The Chicago row shows why the check pays. The model forecast a high of 86.2. The market put 99% on the 88 to 89 bracket and priced the model's 86 to 87 bracket at one cent [13]. Midway had already recorded 88 [14], 1.8 degrees above the model [17]. The author's verdict: "The market wasn't confident, it was informed; the model run was old." [15]
What to watch
- Whether Kalshi corrects the KXHIGHTSDF and KXLOWTSDF titles in the series list.
- Whether series or event objects gain a structured station or city field that replaces parsing the rules text.
- A count across all 422 entries of titles that disagree with their rules, which would size a problem the post shows in two entries.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence58
- Adoption
- Insufficient
- Hype gap+5
- Incentives
- Insufficient
- Confidence60
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
On 12 October 2026 the author called the public series endpoint filtered to the Climate and Weather category and got 422 entries, two of which the post prints.
- [2]
In that list, the series with ticker KXHIGHTSDF is titled 'HIGHEST Temperature SATX'. SDF is the IATA code for Louisville; SATX is San Antonio.
- [3]
San Antonio has its own series, KXHIGHTSATX, titled 'San Antonio Daily Maximum Temperature'.
- [4]
The event KXHIGHTSDF-26OCT12 under the same series is titled 'Highest temperature in Louisville on Oct 12, 2026?'.
- [5]
The event's rules say that if the maximum temperature recorded at Louisville (CLISDF) for Oct 12, 2026 is less than 82 degrees Fahrenheit according to The Weather Company, the market resolves to Yes.
- [6]
The sibling series KXLOWTSDF is titled 'Lowest Temperature SDX', which is not a code for anywhere.
- [7]
"The title is a fallback label and nothing else."
- [8]
The author's stationFromRules function matches a parenthesised string of three capital letters followed by two to four capitals or digits in the rules text, returns it, and returns null when nothing matches; the city comes from a station table, never from the title.
- [9]
Newark and Trenton are their own markets, KXHIGHTEWR and KXHIGHTTTN, so 'New York weather' is three separate ladders.
- [10]
These markets settle on one climate report from one point on the map, not on 'the weather in Chicago'.
- [11]
Kalshi's climate-report code maps to the NWS station id by swapping the CLI prefix for K (CLIMDW to KMDW), and api.weather.gov/stations/KMDW/observations gives what the settling station has recorded, free.
- [12]
Settlement is according to The Weather Company per the rules, so the author treats NWS observations as a sanity check on the same thermometer, not as the official number.
- [13]
In the author's Chicago row for 12 October 2026 (station CLIMDW), the model forecast a high of 86.2, the market favourite was the 88 to 89 bracket at 99%, and the bracket the model pointed at (86 to 87) was priced at one cent.
- [14]
Midway had already recorded a high of 88 when the author pulled the data.
- [15]
"The market wasn't confident, it was informed; the model run was old."
- [16]
The station-code pattern accepts strings of 5 to 7 characters.
- [17]
The observed high of 88 was 1.8 degrees above the model forecast of 86.2.
- [18]
The author compared forecasts from one provider, Open-Meteo, for two coordinate pairs per city, daily high for the next three days, pulled on the afternoon of 12 October 2026.
- [19]
Phoenix and Washington stayed inside half a degree; the author says the gap is coastal and lake-effect geography and is biggest exactly where the volume is.
- [20]
Kalshi's brackets are two degrees wide, and the author says a 4.2 degree error is two whole brackets.
Sources
1 independent publisher whose own reporting we read for this story.
- dev.toKalshi's Market Titled 'HIGHEST Temperature SATX' Settles in Louisville
1 article · October 12, 2026
Topics and entities
Follow any of these and your For You feed starts watching them — no settings page required.
Topics
- Weather forecasting dataFollow
- Prediction MarketsFollow
- Market data APIsFollow