In the last 100 years, agricultural developments have favoured selection for highly productive crops, a fact that has been commonly associated with loss of key traits for environmental stress tolerance. We argue here that this is not exactly the case. We reason that high yield under near optimal environments came along with hypersensitization of plant stress perception and consequently early activation of stress avoidance mechanisms, such as slow growth, which were originally needed for survival over long evolutionary time periods. Therefore, mechanisms employed by plants to cope with a stressful environment during evolution were overwhelmingly geared to avoid detrimental effects so as to ensure survival and that plant stress “tolerance” is fundamentally and evolutionarily based on “avoidance” of injury and death which may be referred to as evolutionary avoidance (EVOL-Avoidance). As a consequence, slow growth results from being exposed to stress because genes and genetic programs to adjust growth rates to external circumstances have evolved as a survival but not productivity strategy that has allowed extant plants to avoid extinction. To improve productivity under moderate stressful conditions, the evolution-oriented plant stress response circuits must be changed from a survival mode to a continued productivity mode or to avoid the evolutionary avoidance response, as it were. This may be referred to as Agricultural (AGRI-Avoidance). Clearly, highly productive crops have kept the slow, reduced growth response to stress that they evolved to ensure survival. Breeding programs and genetic engineering have not succeeded to genetically remove these responses because they are polygenic and redundantly programmed. From the beginning of modern plant breeding, we have not fully appreciated that our crop plants react overly-cautiously to stress conditions. They over-reduce growth to be able to survive stresses for a period of time much longer than a cropping season. If we are able to remove this polygenic redundant survival safety net we may improve yield in moderately stressful environments, yet we will face the requirement to replace it with either an emergency slow or no growth (dormancy) response to extreme stress or use resource management to rescue crops under extreme stress (or both).