In the context of the double impacts of climate change and strong human intervention on the water resources system, showing the system’s evolution process and predicting the future is necessary, which plays an important role in evaluating the implemented water resources management measures, proposing future measures, and promoting the sustainability of water resources. This paper explored to apply synergetic theory and entropy theory in the study of the evolution of water resources. Considering water resources as a dissipative system comprising a resource subsystem, social subsystem, economic subsystem, ecological subsystem, and environmental subsystem, this paper proposed a new method called system evolution entropy through optimization selection of order parameters and order degree calculation. Using Beijing as a case study, we quantitatively showed the evolution process of a water resources system from 2000 to 2013 and predicted the evolution direction in recent years. The results indicate that Beijing’s water resources system oscillated slightly with a non-persistent direction in 2000-07, but evolved toward order continuously after the 2008 Summer Olympic Games. Good consistency between the changes of system evolution entropy and the actual changes of Beijing’s water resources system, and socio-economic and ecological environments through comparative analysis demonstrated that the study results fit the actuality well, i.e. the changes of system evolution entropy reveal the evolution process well, thereby indicating that this method exhibits good feasibility, adaptability, and effectiveness. Additionally, Beijing’s water resources system will continue evolving toward order over the next few years by the prediction according to this method.