Few studies have evaluated the effects of alcohol consumption on the incidence of the metabolic syndrome (MetS). Therefore, the objective of the present study was to examine the association between alcohol consumption and incident MetS in a population of US men. This is a prospective study of 7483 Caucasian men, who were free of the MetS and CVD at baseline. Information was collected on alcohol consumption, health status and fitness level at an initial clinical examination. Additional health information and determination of incident cases of the MetS were obtained at follow-up clinical examinations between 1979 and 2005. Compared with non-drinkers, the multivariate hazard ratios of the MetS for light (1–3 drinks/week), moderate (4–7 drinks/week), moderate–heavy (8–13 drinks/week) and heavy ( ≥ 14 drinks/week) drinkers were 0·81 (95 % CI 0·68, 0·95), 0·68 (95 % CI 0·57, 0·80), 0·70 (95 % CI 0·59, 0·83) and 0·78 (95 % CI 0·66, 0·91), respectively. This association was seen across age groups, in men with one or more pre-existing MetS risk factors, and those with BMI ≥ 25 kg/m2, and in all alcohol beverage types at most levels of alcohol consumption. An inverse dose–response association between alcohol consumption and low HDL concentrations was observed, while significant associations were observed between high fasting glucose concentrations and moderate, moderate–heavy and heavy levels of alcohol consumption. Alcohol consumption was not significantly associated with central obesity, hypertriacylglycerolaemia or hypertension. All levels of alcohol consumption provided significant inverse associations with incidence of the MetS. In particular, this effect was observed in overweight and/or obese individuals, in those who had pre-existing risk factors for the MetS, and extended across all types of alcoholic beverages consumed.