Aneuploidies, defined as whole-arm or whole-chromosome imbalances, are the most prevalent alteration in cancer genomes. However, the extent to which they are enriched due to selection is unclear, against the alternative hypothesis that they are passenger events that are simply highly prone to occur. We developed a novel method, BrISCUT, that identifies loci under selective advantage or disadvantage due to arm-level copy-number alterations by interrogating length distributions of events that are bounded at either the telomere or centromere. These loci were significantly enriched for known cancer driver genes, including genes not detected through analysis of focal copy-number events, and were often lineage-specific. We also formally quantified the role of selection and mechanistic biases in driving aneuploidy, finding that rates of arm-level SCNAs are most highly correlated with selective pressures. These results provide insight into the causes of aneuploidies and their contributions to tumorigenesis.