Short tandem repeats (STRs) are ubiquitous throughout the human genome and routinely vary within human populations but have largely been excluded from genome-wide analyses of variant contributions to human phenotypes. In my thesis work, I and my collaborators demonstrate that current bioinformatic advances now allow for the inclusion of STRs in such studies. We present evidence suggesting that STRs are likely causal for 5.2-7.6% of signals for human blood traits as well as making widespread impacts on gene expression across different tissues. We demonstrate how to carefully interpret and maximize the reliability of statistical fine-mapping to overcome high degrees of correlation between nearby variants, showing its central utility for the study of complex traits. So doing, we uncover many putatively causal STRs strongly affecting human phenotypes.