Human seasonal influenza viruses cause significant morbidity and mortality on a global scale. The constant evolution and seasonal epidemic transmission of these viruses has allowed the virus to continually evade human population immunity and produce novel strains. To develop effective preventative measures, it is critical to leverage available data and algorithmic frameworks to accurately characterize and track virus evolution. This body of work describes methods that aim to robustly characterize influenza virus evolution on the national and international scale. To perform this characterization each chapter interrogates and introduces different statistical methodologies for the study of influenza. In the first aim of this thesis, phylogeographic methods were employed to study the nature of viral diffusion in the United States. A data informed geographic partitioning schema was used to developed and leverages in jointly estimated discrete trait diffusion models over a decade of H3N2 influenza transmission. This work identified major geographic sources and sinks for influenza across seasonal epidemics and identified important predictors for the transmission process. The second aim of this work combines Bayesian phylogenetic methods with antigenic cartographies inferred using inhibition assay data in a generalized additive model to study the influence of different predictors, such as climate and demographic information, on the evolutionary landscape for multiple seasonal influenza viruses (H3N2, H1N1, B-Yamagata, and B-Victoria). This work introduces a novel methodology for using phylogenetic metrics as a response variable to study the partial effects of antigenic space on virus evolution. In the third and final aim of this work antigenic cartographies for H3N2 viruses are compared to assess the assays they are derived from. Antigenic cartography methods use log transformed titer data, traditionally hemagglutinin inhibition assay titer data, to study the differences between isolates of influenza. Newer neutralization-based assays have been introduced to address recent changes in HI assay sensitivity. This chapter compares the ordinations made by different assays as well as with phylogenetic distances and shows a stronger correlation between neutralization-based assays as well as low correlation with phylogenetic history across assays. Overall, this thesis provides a framework to make actionable inferences about influenza transmission and evolution.