This study compares two commonly used rainfall–runoff models in Indonesia, namely the F.J. Mock and NRECA models, for estimating river discharge in three main river basins (DAS) in West Java: Cisangkuy, Cibeet, and Ciwidey. The limited availability of observed discharge data has prompted the use of models as an alternative for discharge estimation in water resource planning. The F.J. Mock model is based on a monthly water balance, while the NRECA model is an empirical model that relies on soil moisture and daily rainfall parameters. The data used include rainfall from six stations and discharge data for the period 2013–2023, analyzed using HEC–HMS, CLIMWAT, and CROPWAT software. Rainfall data validation was performed using the Double Mass Curve and RAPS consistency tests, followed by discharge calculations using both models. Performance evaluation was conducted based on the Nash–Sutcliffe Efficiency (NSE), coefficient of determination (R²), and Percent Bias (PBIAS). The results indicate that the NRECA model performs better, with higher NSE and R² values and lower PBIAS compared to the F.J. Mock model. Additionally, NRECA is considered more practical as it requires fewer data inputs while still producing accurate discharge estimates. Therefore, the NRECA model is recommended for use in watersheds with limited data, while the F.J. Mock model is more suitable for long-term analysis in watersheds with complete data. This study emphasizes the importance of selecting the appropriate rainfall–runoff model to support effective and adaptive water resource management.