Intense training results in numerous physiological perturbations such as muscle damage, hyperthermia, dehydration and glycogen depletion. Insufficient/untimely restoration of these physiological alterations might result in sub-optimal performance during subsequent training sessions, while chronic imbalance between training stress and recovery might lead to overreaching or overtraining syndrome. The use of post-exercise cold water immersion (CWI) is gaining considerable popularity among athletes to minimize fatigue and accelerate post-exercise recovery. CWI, through its primary ability to decrease tissue temperature and blood flow, is purported to facilitate recovery by ameliorating hyperthermia and subsequent alterations to the central nervous system (CNS), reducing cardiovascular strain, removing accumulated muscle metabolic by-products, attenuating exercise-induced muscle damage (EIMD) and improving autonomic nervous system function. The current review aims to provide a comprehensive and detailed examination of the mechanisms underpinning acute and longer term recovery of exercise performance following post-exercise CWI. Understanding the mechanisms will aid practitioners in the application and optimisation of CWI strategies to suit specific recovery needs and consequently improve athletic performance. Much of the literature indicates that the dominant mechanism by which CWI facilitates short term recovery is via ameliorating hyperthermia and consequently CNS mediated fatigue and by reducing cardiovascular strain. In contrast, there is limited evidence to support that CWI might improve acute recovery by facilitating the removal of muscle metabolites. CWI has been shown to augment parasympathetic reactivation following exercise. While CWI-mediated parasympathetic reactivation seems detrimental to high-intensity exercise performance when performed shortly after, it has been shown to be associated with improved longer term physiological recovery and day to day training performances. The efficacy of CWI for attenuating the secondary effects of EIMD seems dependent on the mode of exercise utilised. For instance, CWI application seems to demonstrate limited recovery benefits when EIMD was induced by single-joint eccentrically biased contractions. In contrast, CWI seems more effective in ameliorating effects of EIMD induced by whole body prolonged endurance/intermittent based exercise modalities.