Countless studies have performed event-based hyetograph-hydrograph analyses to gain insights on runoff and streamflow generation in response to precipitation events. However, most of these studies focused on small, experimental forested catchments, thereby neglecting larger watersheds with heterogeneous land use and various degrees of human impact. Furthermore, event-scale studies are largely biased toward rainfall-runoff analyses, thereby ignoring snowmelt and rain-on-snow events susceptible to generate significant flow in Northern latitudes. The goal of the present study was, therefore, to examine runoff and flow generation in contrasted watersheds, by (1) documenting hydrologic responses to individual rain-only, snowmelt-only, and rain-on-snow events, and (2) assessing the predictability of hydrologic response from landscape characteristics. The focus was on the Lake Erie Basin (LEB), specifically on 99 watersheds (1.8 to 16,410 square kilometers) located in the Canadian and American portions of the LEB. Daily gridded climate data and gauged streamflow data spanning 2000-2019 were used, resulting in the identification and analysis of 28,123 precipitation-runoff events. Results show that rainfall events were rarely identified as the triggers of the largest runoff responses. Conversely, snowmelt and rain-on-snow events led to high-magnitude and flashy runoff responses for 70 out of 99 watersheds. Watershed characteristics pertaining to topography, soil, and land use were good predictors of the temporal variability of runoff response metrics. Given the link between runoff-driven nutrient mobilization and recurrent algal blooms in Lake Erie, this study offers a comprehensive assessment of the variability of runoff responses and triggering precipitation events across the large transboundary basin.