Sizing the nZEB systems properly is crucial for nZEBs to achieve the desired performances. The energy demand prediction uncertainties and the components’ degradation are two major factors affecting the nZEB systems sizing. The energy demand prediction has been studied by many researchers, but the impacts of degradation are still neglected in most studies. Neglecting degradation may lead to a system design that can perform as expected only in the beginning several years. This paper, therefore, proposes an uncertainty-based life-cycle performance analysis (LCPA) method to study the impacts of degradation on the nZEBs longitudinal performance. Based on the LCPA method, this study also proposes a two-stage method to enhance the nZEB system sizing. The study can enhance the designers’ understanding of the components’ degradation impacts. Case studies show that an nZEB might not achieve zero energy targets after years due to degradation. The proposed two-stage design method can effectively mitigate this problem.