Given the mismatch between the large volume of data archived for the sixth phase of the Coupled Model Intercomparison Project (CMIP6) and limited personnel and computational resources, only a small fraction of the CMIP6 archive can be downscaled. In this work, we develop an approach to robustly sample projected hydroclimate states in CMIP6 for downscaling to test whether the selection of a single initial condition (IC) ensemble member from each CMIP6 model is sufficient to span the range of models over the conterminous United States (CONUS) and CONUS sub-regions. We calculate the pattern-centered root mean square difference of IC ensemble members relative to multi-model ensemble averages for shared socioeconomic pathway (SSP) projections over 30-year time periods and compare the ratio of inter-model to intra-model variability for this metric. Regardless of SSP, inter-model variability is much greater than intra-model variability at both the scales of the CONUS as a whole and CONUS sub-regions, indicating that selecting a single IC ensemble member per model is sufficient to sample the range of projected hydroclimate states in the 21st Century. Regionally-resolved Taylor diagrams identify where more IC ensemble member downscaling efforts should be focused. Our results suggest that, with parsimonious sampling, the cost of downscaling temperature and precipitation fields over the CONUS for CMIP6 may have increased only marginally over previous CMIP activities in spite of greatly increased data volumes.