Rainfall-Runoff models, such as Hydrologic Modeling System (HEC-HMS), Stormwater Management Model (SWMM), etc., are used for predicting the hydrologic response of watersheds. An issue that to be addressed by users of these models is the estimate of an appropriate level of watershed subdivision for simulating runoff. The size and number of subwatersheds can affect a watershed modeling process and subsequent results. The research reported herein is exploration to evaluate the effect of various levels of watershed subdivision on simulated runoff hydrographs. The HEC-HMS program was applied to five watersheds in Central Texas with drainage areas that varied from 12.3 to 166 square miles. Each of the five watersheds is near one of the following urban areas: Austin, Dallas (the rural watershed studied is close to Dallas), Fort Worth, San Antonio. In this study, the models were intentionally left uncalibrated, thus the subdivision specification is the result of application of engineering hydrologic modeling practice, as would occur when modeling ungaged watersheds. Watershed subdivision was performed using an iso-area arrpoach where each contributing subdivision contained roughly the same area. The results of the HEC-HMS analysis indicated that the number of subwatersheds had little effect on computed runoff hydrographs. No consistent pattern emerged as to whether lumped or multiple subbasins produce superior results; thus, the appropriate level of subdividing a watershed is still unanswered by this study.