The significance of quantitative risk estimates depends not only on the number of accident events included, but also on the criteria differentiating accidents and accident precursors. Based on exemplary data from the European Commission＇s Major Accident Reporting System, the conventional risk estimator associated with the frequency of several types of major accidents is quantified. Bayesian calculations show that incorporating accident precursors in the accident frequency estimation results, under certain conditions, in significantly increased frequency estimates.