This article investigates the time-varying group formation tracking problems for multiagent systems where the agents are divided into multiple groups and each one achieves different goals. Specially, competition is allowed between different subgroups and there exists cooperation within each group. On this premise, distributed Nash equilibrium seeking strategy is utilized to search for the optimal relative evolutionary trend of each group, and formation tracking control protocol is designed to enable followers in each group to track their leader. Moreover, considering that the velocity signals are usually not available in practical situations, Nash equilibrium seeking strategy and formation tracking control are modified without velocity measurements. According to Lyapunov-based theory, it is proven that both the convergence of the Nash equilibrium seeking between subgroups and the convergence of formation tracking error within each group can be fulfilled. A simulation is provided to demonstrate the theoretical results.