We report a tool for analysing through statistical model checking, complex dynamical systems on graphs that can be modelled as multi-agent systems. We discuss techniques to leverage the fact that we restrict the tool to dynamics on graphs for performance improvements. The query language that the tool provides is a probabilistic version of bounded linear temporal logic. We also introduce the notion of population sampling on agents for statistical model checking. To the best of our knowledge, this feature has not been reported previously in literature. Finally, we report experimental results on running examples that illustrate our ideas and the utility of the tool. © 2019, Springer Nature Switzerland AG.