Performance analysis, which plays a key role in the design stage, is employed to estimate whether product performance can satisfy design requirements. In general, product performance is gained after parts are assembled; product performance is influenced by the position and orientation deviations (PODs) that occur in directions of the constrained degrees of freedom (DOFs) due to the surface deviations of mating-surfaces. Furthermore, PODs are uncertain because the surface deviations as well as positions in the unconstrained DOF directions can vary randomly. Thus, predicting the consequences of uncertain PODs on product performance is key for performance analysis. Considering that planes are extensively used in assemblies, this study aims to propose a statistical approach to analyze the uncertain PODs of non-ideal planes. A modeling method from the perspective of manufacturing errors is employed to describe the uncertain surface deviations. A method for computing the uncertain PODs based on the progressive adjustment of coordinate systems is proposed. The maximum PODs that characterize the most unfavorable assembly situation are determined as evaluation indicators. Finally, the effectiveness of the presented approach is verified by a case study. Because both the effects of uncertain surface deviations and uncertain positions on PODs can be considered, the approach is expected to help predict the practical effects of uncertain PODs on product performance accurately during the design stage.