We examine a repeated one-to-one two-sided decentralized matching environment. Using an experimental strategic framework, we seek to characterize the convergence outcome of this process and its sensitivity to environmental factors such as the information structure.