We present the design and evaluation of HyperStorylines, a technique that generalizes Storylines to visualize the evolution of relationships involving multiple types of entities such as, for example, people, locations, and companies. Datasets which describe such multi-entity relationships are often modeled as hypergraphs, that can be difficult to visualize, especially when these relationships evolve over time. HyperStorylines builds upon Storylines, enabling the aggregation and nesting of these dynamic, multi-entity relationships. We report on the design process of HyperStorylines, which was informed by discussions and workshops with data journalists; and on the results of a comparative study in which participants had to answer questions inspired by the tasks that journalists typically perform with such data. We observe that although HyperStorylines takes some practice to master, it performs better for identifying and characterizing relationships than the selected baseline visualization (PAOHVis) and was preferred overall.