Two distinct types of statistical models are used in medical research. Etiologic models examine a potential causal association between an exposure and an outcome (typically while controlling for confounding variables). Predictive models aim to predict the individual risk of an outcome using multiple covariates that may or may not have a causal association. Both models are useful in surgery, where individualized risks can be used to inform surgeons, patients, and their families about risks of perioperative outcomes. For example, the American College of Surgeons National Surgical Quality Improvement Program Surgical Risk Calculator is a widely adopted, web-based decision aid that uses estimates from a risk prediction model to inform surgeons and patients about the estimated risk of 30-day postoperative complications.¹ However, not all models follow high standards before dissemination in peer-reviewed publications or incorporation into decision aids.