Hindsight bias occurs when people overestimate what was known before an event once the outcome is known. This work describes an application of a specific cognitive model, Reconstruction After Feedback with Take the Best (RAFT), as a computational model. It takes the form of a supervised classification model utilizing K-Nearest Neighbors, Decision Tree, and Random Forest machine learning techniques, and is implemented in two settings, novel COVID-19 and everyday nutrition. Testing reveals outcomes are dependent on the domain. Nutrition is found to be associated with greater hindsight bias than COVID-19, and COVID-19 deaths to be associated with greater hindsight bias than cases. The results are discussed with regard to their potential applications for hindsight bias literature and the field of machine learning. The results are discussed with regard to their potential applications for hindsight bias literature and the field of machine learning.