Many observational studies have shown an association between moderate alcohol consumption and a lower risk for cardiovascular morbidity and mortality. Some of these studies, whether or not inspired by the French paradox, suggest a more favourable effect of wine than of other alcoholic drinks. Certain polyphenols including the flavonoids, more abundant in red than in white wine, are held responsible for this 'bonus' effect. However, this conclusion seems premature, since no significant bioactive effect of wine polyphenols has been shown in humans so far. Furthermore, wine drinking proves to be associated with a healthier lifestyle profile than consumption of beer and liquor, and this may have a substantial influence on the outcome of studies. In contrast to moderate drinking, incidental heavy or binge drinking is associated with an increased cardiovascular risk by influences both on the electrical conduction system of the heart and the process of atherothrombosis. Although only prospective randomised intervention trials including a sufficient number of people will give definite answers, the chances are small that they will ever be performed given the ethical and practical objections of such studies. Available data so far justify the conclusion with regard to cardiovascular risk that the pattern of drinking is of more importance than the content of the bottle.