Defining a suitable erosive factor for interrill erosion has been a long-going discussion that was not resolved by experimental results.: Insights about the choice of an erosive factor. In this paper, by using computational fluid simulations of individual raindrop impacts, the relevance of four erosive factors was assessed for a range of soil resistance, rainfall type and water layer depth. Computation results show that the erosivity exponent associated to the erosive factors had a very low sensitivity to soil resistance. This confirms the common practice of separating rainfall erosivity from soil erodibility. The erosivity exponent was found to be very sensitive to the rainfall type. Therefore, an universal exponent could not be found, its value depending on the rainfall type. As a result of the balance between the shear stress development and the protective effect of the water layer, the soil detachment rate was maximum for a specific water depth. This supports to better account for the water layer depth in erosion models. Overall, the computational approach shows that the choice of an erosive factor may not be of practical importance. It also encourages for new experimental designs, able to evaluate soil detachment by raindrops — and not only the amount of splashed soil.