The SOD1G93A mouse has been used since 1994 for preclinical testing in amyotrophic lateral sclerosis (ALS). Despite recent genetic advances in our understanding of ALS, transgenic mice expressing mutant SOD1 remain the best available, and most widely used, vertebrate model of the disease. We previously described an optimised and rapid approach for preclinical studies in the SOD1G93A mouse. Here we describe improvements to this approach using home cage running wheels to obtain daily measurements of motor function, with minimal intervention. We show that home cage running wheels detect reductions in motor function at a similar time to the rotarod test, and that the data obtained are less variable allowing the use of smaller groups of animals to obtain satisfactory results. This approach refines use of the SOD1G93A model, and reduces the number of animals undergoing procedures of substantial severity, two central principles of the 3Rs (replacement, reduction and refinement of animal use in research). The small group sizes and rapid timescales enable affordable large-scale therapeutic pre-screening in the SOD1G93A mouse, as well as rapid validation of published positive effects in a second laboratory, one of the major stumbling blocks in ALS preclinical therapy development.