The publication of results from recent cardiovascular outcome trials (CVOTs) has transformed the landscape of diabetes treatment. Glucagon-like peptide-1 receptor agonists (GLP-1RAs) and sodium glucose co-transporter-2 (SGLT2) inhibitors have demonstrated CV benefits in large, well-conducted, randomized studies. Today, empagliflozin, canagliflozin and liraglutide are US Food and Drug Administration-approved not only for glucose-lowering, but also to reduce the risk of cardiovascular (CV) events/CV mortality in people with type 2 diabetes (T2DM) and established CV disease (CVD)/high CVD risk. Although the CVOTs were primarily powered for CV safety (non-inferiority), they also demonstrated CV efficacy (superiority). This initially surprised many in the diabetes community, but the replication of the CV benefits with different compounds in the same class alleviated concerns about the CV benefits being chance findings. However, many questions remain. While the heterogeneity in the CV benefits in the various CVOTs can be attributed to the variability in CV risk in the different studies, the reason(s) for the differences in the CV benefits between the GLP-1RA class and the SGLT2 inhibitor class appear to be more complex. An analysis of major adverse cardiovascular events (MACE) in the CVOTs shows that the CV benefits of GLP-1RAs are predominantly specific to atherosclerotic CV events (non-fatal myocardial infarction [MI], non-fatal stroke and CV death). By contrast, the SGLT2 inhibitors do not have any significant effects on atherosclerotic CV events (non-fatal MI/stroke). Their benefits are predominantly on hospitalization for heart failure (HF), suggesting effects primarily on myocardial function ("the pump"), and not on the "pipes" (coronary arteries). In the present review, we discuss the rationale for the conduct of CVOTs, highlight the inability of the classic three-point MACE to capture the entire spectrum of atherosclerotic and non-atherosclerotic CVD morbidity, especially HF in T2DM, and discuss the results of the CVOTs with a focus on the clinical significance of atherosclerotic CVD (ASCVD) versus HF, which develops in a sizeable proportion of people with diabetes and without prior ASCVD.