The optimal tracking performance of networked control systems with communication delay and white Gaussian noise under the consideration of the channel input power constraint is studied in this paper. The tracking performance is measured by the energy of the system’s error signal and channel input power. The tracking performance is minimized by considering the channel input power constraint searching through all stabilizing two-parameter controllers. The optimal tracking performance of networked control systems under the communication delays, packet dropout and white Gaussian noise of the communication channel has been studied. The explicit expressions of the optimal tracking performance are obtained by applying the spectral factorization and inner–outer factorization techniques. It is shown that the optimal tracking performance depends on the non-minimum phase zeros, unstable poles of the given plant, communication delay, packet dropouts probability and white Gaussian noise. It is also shown that if the constraint of the communication channel does not exist, the optimal tracking performance reduces to the existing tracking performance of the control system without communication. The result shows how the communication delay, packet dropout probability and white Gaussian noise may fundamentally constrain a networked system’s tracking capability. Finally, some typical examples are given to illustrate the theoretical results.