Twenty years ago it was argued that rotational wheat production systems will reduce the economic risks to farmers and restore soil quality. Here we reflect on this assertion by analysing the evidence of a 12-year data window within a trial on a mixture of crop rotation systems at Langgewens Research Farm, South Africa. It was been found that production systems that include rotations with medics and/or medic-clover show some potential for improvement compared with wheat only, with a combination of the annual legume pasture with an added saltbush pasture showing the greatest improvement when taking into consideration the benefits from livestock production that are derived from pastures. Pastures are more resilient to changes in rainfall compared with wheat only. Planting pastures in alternate years also improves the yields from wheat, and this is beneficial in periods of low rainfall. Rotation systems on this farm that include lupin perform worse than the wheat-only model. Furthermore, when modelling the effect of drought on the system, the results of the multi- and rotation production systems actually improve.