Climate-flood relationships in the midcontinental United States (US) are not well understood prior to instrumental climate and stream discharge records because there are limited paleoclimate records capable of distinguishing seasonal climate variations and continuous paleo-flood records from watershed of different scales are sparse. Characterizing long-term climate flood relationships is critical to determine “natural” climate-flood dynamics and identify if and how they were impacted by anthropogenic land use (e.g., deforestation, conversion to farmland, wetland drainage, and urbanization) and climate changes. Here, paleoclimate data from Martin Lake, Indiana, and paleo-flood reconstructions from streams with large (> 100,000 km²) and medium to small (< 100,000 km²) watersheds in the midcontinental US spanning the Common Era (CE; last 2000 years) are synthesized. The data show that flooding on streams with large watersheds was driven by changes in continental snowpack with reduced flooding prior to and during the Medieval Climate Anomaly (MCA; 950 to 1250 CE) when US snowpacks were diminished and increased flooding during the Little Ice Age (LIA; 1250 to 1830 CE) when US snowpacks increased. Small streams in the eastern midcontinent on the other hand experienced increased flooding during the MCA when rainstorms were more frequent and decreased during the LIA when cold-season-like conditions predominated rainstorms diminished. These climate-flood dynamics were driven by mean state changes in synoptic-scale atmospheric circulation resembling the Pacific North American mode (PNA), with negative PNA-like conditions enhancing rainstorm events and diminishing snowpacks during the MCA and positive PNA-like conditions reducing rainstorms events and enhancing snowpacks during the LIA. Steams in watersheds of all sizes show increased flooding during the last 150 years despite predominantly negative PNA-like conditions, indicating that recent anthropogenic land use changes have contributed to flooding and erosion by reducing infiltration and enhancing runoff.