Kang and Lansey (2008a) have shown that water distribution system (WDS) nodal demands can be estimated in real-time using pipe velocity measurements from a supervisory control and data acquisition (SCADA) system. The field measurements are key elements for the real-time state estimation. However, the limited number of metering locations has been a significant obstacle for the real-time studies and identifying locations to best gain information is critical. Previous studies for the data sampling mainly focused on minimizing either parameter or prediction uncertainties. However, reducing uncertainty does not guarantee a good fit for the model predictions in terms of the mean estimate. Therefore, robust objective criteria, that guarantee precise and accurate state estimates, must be applied. Here, an optimal meter placement (OMP) problem is formulated as a multi-objective optimization model. Three distinctive objectives are posed: (1) minimization of nodal demand estimation uncertainty; (2) minimization of nodal pressure prediction uncertainty; and (3) minimization of absolute error between demand estimates and their expected values. Objectives (1) and (2) represent model precision while objective (3) describes model accuracy. The OMP problem is solved using a multi-objective genetic algorithm (MOGA) based on Pareto-optimal solutions. The trade-off between model precision and accuracy is clearly observed from a simple network study and it is strongly recommended to use both criteria as objectives.