Kalshi vs Polymarket weather markets, compared
How daily-temperature markets differ across Kalshi and Polymarket: resolution source (NWS CLI vs Weather Underground), stations, time windows, brackets, fees and regulation — and exactly where a 1°F gap flips a contract.
Kalshi and Polymarket both run daily-temperature markets, and on the surface they look interchangeable. They are not. They use different data sources, different stations, different time windows and different bracket structures — which means the same weather can produce different outcomes on each. That’s a trap if you’re careless and an opportunity if you’re deliberate.
Side by side
| Dimension | Kalshi | Polymarket |
|---|---|---|
| Settlement source | NWS Climate Report (CLI), next day | Weather Underground, day-of |
| Time window | Local Standard Time (shifts in DST) | Local midnight to midnight |
| Example NYC station | Central Park (KNYC) | LaGuardia (KLGA) |
| Bracket width | ~6 brackets, middle four ~2°F | Fine-grained 1° buckets |
| Structure | Independent YES/NO per bracket | Neg-risk multi-bucket event |
| Regulation / rails | CFTC-regulated, USD | Crypto (USDC), on-chain |
| Geography | US cities | 35+ global cities |
| Fees | Per-contract fee (brutal on cheap tails) | Spread-driven; gas on-chain |
The 1°F that flips a contract
The decisive difference is the settlement feed. Kalshi reads the official NWS Climate Report, which can round or report a degree higher or lower than Weather Underground’s figure for the same station and day. When the high lands near a bracket boundary, that single degree can settle a Kalshi bracket YES and the equivalent Polymarket bucket NO. New traders get caught by this; sharp ones watch both feeds on purpose.
Daylight saving breaks naive assumptions
Kalshi’s “day” runs in Local Standard Time. During daylight saving, that window effectively shifts to 1:00 AM–12:59 AM local clock time — so a late-night or early-morning extreme can fall on a different trading day than your intuition says. If your model assumes a clean local midnight-to- midnight day (as Polymarket uses), you will occasionally mis-assign the high.
Which should you trade?
- Kalshi if you want CFTC-regulated USD rails and US cities, and you can stomach per-contract fees (mind the cheap-tail death zone).
- Polymarket if you want fine-grained 1° buckets, crypto rails, and — crucially — global cities, where edge survives longer because the bots are sparser.
- Both if you want to exploit the settlement divergence directly: hedge across platforms, or take the side the gap favours.
See both feeds in one place
Temprr is built station- and source-aware: it knows which gauge and feed settle each market on each platform, surfaces both readings with a boundary-divergence flag, and prices a calibrated distribution against the live odds on each. The cross-platform gap stops being a surprise and starts being a signal.