Methods are described for the testing of automotive hub units under quasi-static, impact, and dynamic conditions. It is shown that the overall time required to reliably determine the rolling contact fatigue life of such units can be reduced by the selection of a suitable test strategy. A strategy is described in which fifteen ‘sudden death’ groups of four units are tested simultaneously. When any one unit in a group fails, then the complete group is considered to have failed. Weibull analysis is then applied to these results to estimate the L10 life of the unit. This ‘sudden death’ strategy has a better discriminating power than a conventional ‘five failures from ten tested units’ strategy, and can be completed in a much shorter test time.