ERCOT Hourly Demand + Population-Weighted Weather
- rows
- over 96,000
- 2015 to 2026
- 11 yrs
- columns
- 29
- baseline MAPE
- 4.84%
Grid demand joined to population weighted ERA5 weather, UTC aligned across every daylight saving transition. Every imputed value is flagged in the data, not buried in a footnote. A LightGBM baseline reaches 4.84% MAPE against an 8.72% seasonal naive benchmark, trained through 2024-12-31 and tested from 2025-01-01 onward, roughly 18 months.
Inside the data
29 columns per UTC hour. Demand ships with its provenance: the canonical MWh value, the as reported value, the PUDL imputed value with its reason code, and a demand_source column separating cleaned history from the preliminary tail. Weather is ERA5 at the four major Texas metros, DFW, Houston, San Antonio, and Austin, plus a population weighted temperature built from Census Vintage 2025 weights. Holidays ship in three semantics, statutory federal, observed federal, and federal plus Texas state. Degree hours and degree days use the standard base 65 F, and quality flags mark imputed demand, missing or interpolated weather, and DST transition days. The demand temperature curve is the dataset's signature: a U with its trough near 64 F, cold arm anchored by Winter Storm Uri and hot arm by the August 2024 record peak.
Provenance and limits
The weather model is pinned to ERA5, never the archive's default blend, so rebuilds reproduce the same series and values match the ERA5 citation. The demand backbone is Catalyst Cooperative's PUDL cleaned EIA-930; the freshest roughly 2 months come straight from the EIA API and are flagged preliminary and unimputed. Only interior weather gaps of 3 hours or less are interpolated, and flagged. A machine generated provenance file ships with each build, recording source date ranges, the pinned weather model, and checksums of both data files.
Stated limits: the four metros cover roughly two thirds of Texas population while ERCOT covers about 90 percent of Texas load, and El Paso is deliberately excluded because it sits in WECC, not ERCOT. ERA5 is a reanalysis estimate, not station observations. A backtest on this data uses actual weather, so its error should not be compared head to head with ERCOT's live day ahead forecast, which runs on weather forecasts and analyst correction.
Attribution
This dataset is published by Arimancy LLC under CC BY 4.0.
Demand data:Source: U.S. Energy Information Administration (Jul 2026), Form EIA-930 (public domain), cleaned and gap-filled via Catalyst Cooperative's PUDL project (data and documentation CC BY 4.0).
Weather data: Weather data by Open-Meteo.com (CC BY 4.0), ERA5 reanalysis (Hersbach et al. 2023, ECMWF, doi:10.24381/cds.adbb2d47).