The distribution of load has high uncertainty, which is the main cause of a rack structure’s instabilities. The objective of this study was to identify the most unfavorable and favorable load distributions on steel storage racks with and without bracings under seismic loading through a stochastic optimization—a genetic algorithm (GA). This paper begins with optimizing the most unfavorable and favorable load distributions on the steel storage racks with and without bracings using GA. Based on the optimization results, the failure position and seismic performance influencing factors, such as the load distributions on the racks and at hazardous positions, are then identified. In addition, it is demonstrated that the maximum stress ratio of the uprights under the most unfavorable load distribution is higher than that under the full-load normal design, and it is not the case that the higher the center of gravity the more dangerous the steel storage rack is, demonstrating that the load distribution pattern has a significant impact on the structural safety of steel storage racks. The statistics of the distributions of the load generated during the optimization of the GA and the contours of the probability distributions of the load are generated. Combining the probability distribution contours and the GA’s optimization findings, the “convex” distribution hazard model and the “concave” distribution safety model for a steel storage rack with bracings are identified. In addition, the features of the distribution hazard model and the load distribution safety model are also identified for steel storage racks without bracings.