The industrial experience at Daqing showed that chemical flooding significantly could enhance the oil recovery. Meanwhile, the fluid behaviors in the reservoirs became more complex due to the injection of chemicals. The dynamic connectivity between injection and production wells formed by water flood has been changed significantly due to the participation of chemical solutions. In this research, an improved grey relation analysis method (IGRA) is introduced to predict the dynamic connectivity between injection and production wells and to estimate the effects of connectivity changes. Plenty of sensitivities in chemical floods are considered in this new method so that the influence of errors caused by a poor indices choice in the variable process description can be minimized. An application was carried out in a polymer flooding pilot of the Xingshugang Unit, Daqing. By comparing the results of both the water and chemical flooding prediction results with IGRA, the dynamic connectivity changes have been demonstrated. Tracer tests have been conducted for both water and chemical flood. The interpretations of the tracer tests have also proven the accuracy of predictions. The results of chemical flooding prediction have been used to adjust the injection and production rate for individuals.