Peer-to-peer (P2P) lending, which has received considerable research attention in recent years, has emerged as a viable platform to provide an alternative way of financing without the help of traditional intermediaries such as banks. However, P2P lending platforms have experienced risks in China, and previous research mainly focuses on using subjective methods for constructing and assessing the risk index system of P2P platforms. We propose a novel approach to establishing a risk index system based on previous research and real-world data, selected using correlation analysis and principal component analysis. Moreover, the entropy-revised G1 combination method has been used in this paper to assess the risk performance of P2P lending platforms, with both expert knowledge and real-world data taken into account. In this paper, we use information contribution to show the constructed risk index system is valid. In addition, we rank the ninety-two P2P lending platforms based on their calculated risk evaluation scores, which are further tested with the ranking offered by China’s rating agency using Spearman’s coefficient. Testing results show the proposed risk assessment for P2P platforms in China is rational and efficient, giving us unique insight into how to evaluate the risk performance of P2P platforms, and thus can be used by regulators to better supervise these platforms in China.