The ability of the N-methyl- D-aspartate receptor antagonist ketamine to induce a rapid and sustained antidepressant effect has led to a surge in pre-clinical studies investigating underlying mechanisms and seeking novel treatments. Animal models are key to this research as they can provide a behavioural readout linking underlying mechanisms to clinical benefits. However, quantifying depression-related behaviours in rodents represents a major challenge with the validity of traditional methods such as models of behavioural despair (forced swim test and tail suspension test) a topic of debate. While there is good evidence to support the value of using these behavioural readouts to study the effects of stress, these approaches have largely failed to detect reliable phenotypic effects in other disease models. In this systematic review, we identified publications which had tested N-methyl- D-aspartate receptor antagonists in normal animals using either the forced swim test or tail suspension test. We compared findings for different doses and time points and also drugs with different clinical profiles to investigate how well the outcomes in the rodent model predicted their effects in the clinic. Despite clear evidence that N-methyl- D-aspartate receptor antagonists reduce immobility time and hence exhibit an antidepressant profile in these tasks, we found similar effects with both clinically effective drugs as well as those which have failed to show efficacy in clinical trials. These findings suggest that behavioural despair tests in normal animals do not provide a good method to predict clinical efficacy of N-methyl- D-aspartate receptor antagonists.