Strong research evidence has shown that medical male circumcision significantly reduces heterosexual HIV acquisition among men. However, its effectiveness is enhanced by behavioural factors such as condom use. Currently, little is known of unprotected sex associated with male circumcision (MC) among alcohol-drinking tavern-going men, or whether engagement in unprotected sex may differ between men who have been traditionally circumcised and those who have been medically circumcised. The study sought to determine the relative importance of alcohol consumption and MC as correlates of unprotected sex and to compare the risk of engaging in unprotected sex between traditionally circumcised and medically circumcised tavern-going men from two rural villages in North-West province, South Africa. Data from 314 adult men (≥18 years) were analysed. The men were recruited from four bars/taverns using systematic sampling. They responded to questions regarding their demographic characteristics, alcohol consumption, circumcision status and method (where applicable), and engagement in unprotected sex. Descriptive analyses and bivariate and multivariate logistic regression analyses were conducted. Age, education, relationship status, alcohol consumption and traditional male circumcision (TMC) were independently and significantly associated with unprotected sex. Specifically, probable alcohol dependence and traditional circumcision were independent risk factors for engaging in unprotected sex among tavern-going men. Traditionally circumcised men had a higher risk of engaging in unprotected sex than medically circumcised men. Interventions aimed at reducing alcohol consumption, encouraging protective behaviour among men who have undergone TMC, and increasing condom use are needed in bar/tavern settings. HIV prevention education must be urgently incorporated into TMC programmes.