The estimation of arrival type, an important part of the 1997 "Highway Capacity Manual" (HCM) procedure for estimation of delay and level of service (LOS) at signalized intersections, is considered. The research examines whether arrival type can be reliably estimated by the procedures outlined in the HCM. The objective is to review the impact that arrival type has on delay estimation and to evaluate quantitatively the variance and error of arrival types estimated under controlled conditions. Progression quality issues are discussed, and potential additions and clarifications are explored. To understand the impact that arrival type estimations have on delay results, an experiment was conducted using the HCM delay estimation procedure. Participants were shown video clips of traffic at seven different intersection approaches and were asked to indicate the observed arrival type. The results are categorized by response groups (traffic engineering professionals, high school students, and university students) and approaches and are compared with results calculated by HCM procedure. The results show trends of inconsistent responses when the responses are compared within groups, between groups, and with the calculated arrival types. On the basis of the response trends, several additions to the HCM method are explored. Recommendations are to include the impact of green split percentages, estimation of arrival types for each lane group, and general clarification of instructions for estimation of arrival types. The study results suggest that these changes and clarifications should be considered for incorporation into the HCM delay estimation and LOS procedure for signalized intersections.