These data were compiled so that annual wildfire could be modelled across the sagebrush region in the western United States. Our goal was to understand how wildfire probability relates to climate and fuel conditions across the entire sagebrush region. To do this we developed a statistical model that represents the relationship between annual wildfire probability and a small number of climate and fuel variables. Specifically, created predictions of wildfire probability using a biologically plausible logistic regression model that related wildfire probability to mean temperature, annual precipitation, the proportion summer precipitation (PSP), and aboveground biomass of annual herbaceous plants and perennial herbaceous plants. The biomass variables were used as proxies for fine fuel availability. These data represent annual fire occurrence in 1 km pixels (i.e. did a given pixel burn that year), predicted wildfire probability, as well as the three year running average (i.e. average across the current and previous two years) of climate and vegetation variables. These data were collected across the sagebrush region (the extent of the study area is provided by the cell_number_ids.tif file). The climate and vegetation data were compiled using a existing gridded dataset (Daymet) of daily precipitation and temperature, and vegetation data were summaries of annual estimates of aboveground biomass of annual and perennial herbaceous plants from the Rangeland Analysis Platform (https://rangelands.app/). These data can be used to understand spatial and temporal variability in wildfire occurrence and modelled wildfire probability between 1988 and 2019 and how that variability relates to spatial and temporal variability in climate and vegetation.