High-resolution time series of burned area (BA) derived from Sentinel-2 can advance understanding of the determinants and dynamics of fire by incorporating small fires previously excluded from regional analyses. Here, we assessed the drivers of fire frequency, fire size, and fire seasonality across Southeastern Africa via fine (Sentinel-2 MSI) and moderate (MODIS) spatial resolution data. Twenty-six predictors of ignition patterns, fuel load, flammability, and fire spread—factors that underpin fire frequency, size, and seasonality—were incorporated into machine learning models to evaluate their predictive capacity, relative importance, and directional relationships with fire regime attributes. We found large differences between fine and moderate resolution mapping in the estimates of fire frequency, size, and to a lesser extent, seasonality, with models using Sentinel-2 having better predictive performance. However, the relationships between fire regime attributes and predictors were generally consistent between sensors. We found that high fire frequency was associated with fuel load and seasonality and that interannual stability in land cover, livestock density and human population, was associated with low fire frequency. Fire sizes were generally small in both the high and low extremes of the precipitation and vegetation productivity gradient, and in highly transformed areas. The fraction of fire outside of the fire season was higher under low seasonality conditions and under higher human influence. Our results demonstrate the applicability of existing theory of fire dynamics found on moderate resolution data to new fine scale data while providing nuanced insights into fire regime determinants in increasingly transformed landscapes.