Yield losses occurring at the field level, whether due to plant diseases or abiotic stresses, reveal reduced stability of the crop yield potential. The paper argues that the stability of crop yield potential is a trait with a clear genetic component, which can be successfully selected for at the single-plant level and incorporated into high-yielding cultivars. Two novel selection equations with prognostic power are presented, capable to objectively phenotype and evaluate individual plants in real field conditions in the absence of the masking effects of interplant competition and soil heterogeneity. The equations predict performance at the crop stand through the key concept of coefficient of homeostasis and are equally useful for early generation selection and for nonstop selection within finished cultivars in order to continuously incorporate the adaptive (genetic or epigenetic) responses of plants. Exploitation of adaptive responses acquires particular importance in view of the climate change effects on crop productivity and the changing biotic or abiotic micro-environments. Cotton is used as a case study to highlight the potential of nonstop selection for increasing crop yield and for the gradual build-up of disease resistance. In addition, the paper envisions and proposes the formation of international networks of researchers focusing on specific diseases as, for example, the cereal root-rot or the cotton Verticillium
wilt that will concurrently use the proposed strategy in their respective environments to select for resistant genotypes, while gaining a deeper understanding of the nature of the genetic or epigenetic changes at the phenotypic and genomic levels.