By analyzing the effects of environmental variables on plants, changes in plant distribution as a result of climate oscillations can be studied, which is of great significance to plant protection and management policies. Acer cordatum Pax (Aceraceae) is a non-deciduous tree distributed in valleys and streams in eastern China. Due to the effects of changing climate (warmer, drier conditions) and human impacts, the number of wild individuals of A. cordatum has exhibited a decrease trend, which is in urgent need of protection. In this study, the maximum entropy algorithm (MaxEnt) was used to predict the potential geographical distribution of A. cordatum during the Last Interglacial (LIG), and Last Glacial Maximum (LGM), current, 2060s, and 2080s periods. The model used geographic location information of 337 A. cordatum and six climatic variables. The area under the receiver operating characteristic curves (AUC) of the simulation results were more than 0.95, indicating high accuracy in the simulation result. The mean temperature of coldest quarter, precipitation of warmest quarter, precipitation of driest month, and precipitation of seasonality were important climatic variables influencing the geographic distribution of A. cordatum. Based on the simulate results, the potential distribution areas of A. cordatum experienced a process of expansion and then contraction from LIG to the future. In the future, some potential suitable areas provinces will likely shrink (Guizhou, Fujian, and Anhui), even almost disappear (Chongqing), and the general distribution will trend to transfer in a northeastward direction. It is hoped that this study can provide a theoretical reference for the future protection of A. cordatum.