DFW Live Weather Dashboard

Updated at 2026-08-29 14:46:33 UTC · 09:46 AM CT
Sources: AviationWeather.gov METAR JSON, NOAA HRRR AWS Open Data

Current Conditions — KDFW (Wethr.net live obs)

Observed 2026-08-29 14:40 UTC · 09:40 AM CT

Temperature Bucket Probabilities

Probability of the daily high falling in each 2°F bucket. The daily high is the maximum temperature across all forecast hours within the NWS climate day (midnight-to-midnight Central Time). It uses the near-resolution-adjusted blended forecast (HRRR, bias-corrected) when live METAR obs are available, falling back to the blended forecast otherwise. The high can occur at any hour — including early morning before a cold front. Sigma grows with the forecast hour of the high. Use the date selector to switch between today and tomorrow, and the cycle dropdown within each to look back at what each forecast cycle predicted. Bucket boundaries, alignment, and range are fetched live from the Polymarket US API — they shift day to day (even-aligned on some days, odd-aligned on others) to match the actual market structure.

Multi-Model Forecast (HRRR)

Use the variable dropdown to switch between Temperature, Dew Point, Wind Speed, Wind Gust, and Humidity. Use the cycle dropdown to look back at past forecast cycles. Non-temperature variables show raw model output (no bias correction). Observations appear on the Temperature graph only.

Ladder Trade Signals

Temperature ladder strategy (@AlterEgo_eth): spread $1 across adjacent buckets buying YES shares at equal share-counts. If the temp lands in ANY bucket, one pays out while the others expire worthless. Ladders with positive expected value (model P > market cost) are shown, sorted by Kelly% — the Kelly Criterion combines P(win) with payoff ratio to rank risk-adjusted opportunity. GUAR. PAYOUT ladders (cost < $1) guarantee a payout exceeding $1 if any bucket hits — but you still lose the full $1 if the temp misses all buckets. EV+ ladders profit in expectation based on the model. Not trade execution — informational only. Scale by multiplying the dollar amount.

Database Statistics