Rolling element bearings are widely used machine components. The understanding of their performance has improved tremendously over the past century and simulations taking into account the dynamic behavior of individual bearing components are common practice. Such simulations allow product developers in the bearing industry to further optimize these components through better understanding of the dynamic loads the components are exposed to. However, to do this also requires more detailed knowledge of the actual material properties. Where components were originally developed based on datasheet properties and safety margins, it has now become necessary to understand the effect of process on the performance. In this presentation it is illustrated how fiber orientation, porosity and crystallization kinetics add up to explain knock-down factors used in bearing component analysis. The importance of modeling crystallization at high cooling rates is demonstrated.