In this paper we propose a new concept to prioritize the importance of a link in a directed network graph based on an ideal flow distribution. An ideal flow is the infinite limit of relative aggregated count of random walk agents' trajectories on a network graph distributed over space and time. The standard ideal flow, which is uniformly distributed flow over space and time, maximize the entropy for the utilization of a network. We show that the simulated trajectories of random walk agents would form an ideal relative flow distribution is converged to stationary values. This implies that ideal flow matrix depends only on the network structure. Ideal flow matrix is invariant to scalar multiplication and remarkably it is always premagic. Demonstration of ideal flow to the real world network was fitted into Sioux Falls transportation network.