In this work, a comparative performance analysis of robust data reconciliation strategies is presented. The study involves two procedures based on the biweight function and three estimation techniques that use the Welsh, quasi-weighted least squares, and correntropy M-estimators. The aforementioned functions are selected for comparative purposes because their use in the data reconciliation literature has appeared during the past decade. All procedures are properly tuned to have the same estimation and gross error detection/identification capabilities under the ideal distribution. Different measurement models are systematically taken into account, and results are analyzed considering both performance measures (average number of type I errors, global performance, mean square error) and computational load. The comparative analysis indicates that a simple robust methodology can provide a good balance between those two issues for linear and nonlinear benchmarks.