Rollover research data is a valuable resource in the field of accident reconstruction and automotive safety. This data can provide the accident reconstructionist with information regarding number of rolls versus roll distance and vehicle average deceleration during the rollover event. One source of rollover research data readily available to the general public is the National Highway Traffic Safety Administration's (NHTSA) Crash Test Database which includes dolly rollover test data. This data includes testing performed using the FMVSS 208 dolly fixture and the NHTSA Rollover Test Device (RTD) used from 1985 to 1995. The purpose of this paper is to provide a more thorough analysis of the RTD rollover tests performed for NHTSA which includes a history of the RTD test series, detailed review and analysis of the available data and videos to provide a better understanding of the actual vehicle test dynamics. The analyzed RTD test results are then compared regarding number of rolls, roll distance and vehicle average deceleration during each of the rollover tests with other publicly available rollover tests performed that use other mechanisms of initiation of rollover to determine the usefulness of this data in accident reconstruction. It was found from this review and analysis that with regard to accident reconstruction, the use of the NHTSA RTD rollover tests with respect to vehicle average deceleration, number of rolls and roll distance should be done so with great care and consideration .