Naturally occurring stable water isotope tracers provide useful information for hydrologic model development and calibration. Existing models include varied approaches concerning unsaturated zone percolation mixing (preferential versus matrix flow) and evapotranspiration (ET) partitioning. We assess the impact of unsaturated zone simplifying assumptions when simulating the Shale Hills Watershed, a small (7.9 ha), temperate, forested watershed near Petersburg, Pennsylvania, USA, with a relatively simple model. We found that different model structures/assumptions and parameterizations of unsaturated zone percolation had substantial impacts on the agreement between simulated and observed unsaturated-zone water isotopic signatures. We show that unsaturated zone percolation mixing primarily affects the unsaturated zone delta18O and delta2H during winter and spring and that percolation was best represented as a combination of both preferential and matrix flow. We evaluate the importance and implications related to the partitioning of ET into evaporation and transpiration and demonstrated that incorporation of a plant growth model for ET partitioning substantially improved reproduction of observed hydrologic isotopic patterns of the unsaturated zone during the spring season. We show that unsaturated zone percolation mixing and ET partitioning approaches do not substantially influence stream delta18O and delta2H and conclude that observed streamflow isotopic data is not always a strong predictor of model performance with respect to intrawatershed processes.