An intercomparison of gage and radar-based precipitation estimates was undertaken for the Midwestern United States. Multi-sensor precipitation estimates (MPE) based on the Stage III/IV algorithm developed by the Office of Hydrology / NWS River Forecast Centers, National Weather Service quality-controlled cooperative gage (QC-Coop) data, and gage data from three high density networks in Illinois from February 2002 through August 2005 were examined. Most differences in county-averaged monthly precipitation estimated by QC-Coop and MPE in nine Midwestern states were within +/- 25%, averaging 6%. The difference between gage and MPE monthly values decreased somewhat through the 41-month period of study. Data from three regional gage networks indicated that on a daily basis, averaged MPE and gage network data agreed to within about +/- 25%. Daily MPE values were often lower than the gage values for large precipitation amounts. When examining multiple gages within single MPE grid points, it was found that for very low daily precipitation amounts, MPE grid amounts were generally greater than gage, and for high daily precipitation amounts, MPE amounts were lower than gage amounts.