ABSTRACT Previous tornado climatology research has relied primarily upon a mean value of tornadoes or tornado days fixed on one period to define regions of elevated tornado risk. However, a more comprehensive understanding of the climatological distribution of tornadoes is achieved by analysing the spatial and temporal variability of tornado days. Two methods of analysing tornado days for predefined areas across the continental United States using a refined version of the Storm Prediction Center's tornado data set for 1950–2011 are considered in this study. Statistical analyses of averages and return periods reveal the variability in the record, while consecutive tornado days present an alternative method to assess the seasonal repeatability and risk for historic, multi‐day tornado outbreaks. The results of this research better define the climatological variability of tornado events in addition to offering insight regarding the spatial patterns of locations with the greatest tornado risks, especially those with higher interannual variability.