Wildfires and housing development have increased since the 1990s, presenting unique challenges for fire management. However, it is unclear how the relative influences of housing growth and changing wildfire occurrence have contributed to risk to homes. We fit a random forest using weather, land cover, topography, and past fire history to predict burn probabilities and uncertainty intervals. Then, we estimated risk at 1-km resolution and monthly intervals from 1990 through 2019 by combining predicted burn probabilities with housing density across the Southern Rocky Mountains. We used 3 scenarios to evaluate how housing growth and changes in burn probability influenced risk individually and combined (observed, 1990 housing, and 1990 weather). This data release includes python scripts used for all processing steps and a readme file describing where to acquire original datasets used by the random forest model, instructions for running the python scripts, and descriptions of outputs. Preprocessed model inputs were too large to share. However, raster layers are included for modeled burn probability and risk for the 3 scenarios.