Frequency of alcohol drinking is known to influence cardiovascular risk. However, little is known regarding the relationship between frequency of drinking and metabolic syndrome (MetS). The aim of this study was to determine how frequency of heavy drinking modifies the prevalence of MetS. The subjects were middle-aged male nondrinkers and occasional or regular heavy drinkers (ethanol intake: ≥66 g per drinking day). Odds ratios (ORs) for MetS and each component comprising MetS were calculated with adjustment for age and histories of smoking and regular exercise. ORs versus nondrinkers for MetS defined by the criteria of the National Cholesterol Education Program's Adult Treatment Panel III (NCEP-ATP III) or the International Diabetes Federation (IDF) were significantly higher than the reference level of 1.00 in occasional heavy drinkers (NCEP-ATP III, 1.94 [confidence interval (CI): 1.54 to 2.45]; IDF, 1.97 [CI: 1.56 to 2.49]) and regular heavy drinkers (NCEP-ATP III, 1.48 [CI: 1.19 to 1.84]; IDF, 1.50 [CI: 1.20 to 1.86]). When compared with the reference level, OR versus nondrinkers for large waist circumference was significantly higher in occasional heavy drinkers (1.96 [CI: 1.63 to 2.35]), but not in regular heavy drinkers (1.12 [CI: 0.96 to 1.32]), while OR versus nondrinkers for hyperglycemia was significantly lower in regular heavy drinkers (0.66 [CI: 0.46 to 0.95]), but not in occasional heavy drinkers (0.86 [CI: 0.60 to 1.24]). There is a positive association between heavy drinking and MetS, which is stronger in occasional drinkers than in regular drinkers. This difference may be explained by a positive association between occasional heavy drinking and central obesity and an inverse association between regular heavy drinking and hyperglycemia. The results suggest that heavy drinking, even if occasionally, is a cardiovascular risk factor.