The board · Kalshi temperature ladders · live

Every bracket, every hour.

A robot with no forecast reads every open bracket of Kalshi's daily-high and daily-low ladders once an hour: the bid, the ask, the size resting at the ask, and, from the settlement station's own observations, whether the day has already ruled the bracket out. This is its latest pass, a snapshot the database refreshes every fifteen minutes. Nothing here is a recommendation; it is the market, measured.

loading the board…

the Yes ask, 0–100cthe midpointruled out by the observationsize = contracts resting at the ask · queue = Yes asks resting at 1c

How to read a ladder

A ladder is one city, one metric (the day's high or low), one settlement date: a set of brackets that cover every temperature, one of which will settle Yes. Each rung shows the bracket's Yes ask as a bar across 0–100c with the midpoint marked, the bid and the ask in cents, and the contracts resting at the ask. A struck-through rung is a bracket the station's confirmed running extreme has already cleared by a full degree: the observation says it cannot settle Yes, and what remains on it is the queue, the Yes asks resting at one cent, plus any stale bid at 3c or better that somebody could still sell into. A rung in green has settled.

Source: the ladder logger (agents/kalshi_ladder_logger.py) writes every read to the database; board_views.sql builds this page's view from the latest read of each bracket in the last six hours. Bids, asks and sizes are Kalshi's own public quotes at the moment of the read; the verdict is computed from the settlement station's hourly observations with the method on the method page.