Inspired by the ability of Markov chains to model complex dynamics and handle large volumes of data in Google's PageRank algorithm, a similar approach is proposed here to model road network dynamics. The central component of the Markov chain is the transition matrix which can be completely constructed by easily collecting traffic data. The proposed model is validated using the popular mobility simulator SUMO. Markov chain theory and spectral analysis of the transition matrix are then shown to reveal non- evident properties of the underlying road network and to correctly predict consequences of road network modifications. Preliminary results from possible applications are shown and simple practical examples are provided throughout this article to clarify and support the theoretical expectations.