The aim of this research was to evaluate relationships between the PORCUS classification system and factors affecting boar taint in pork. Intact male pork carcasses (n = 176) were randomly selected in a South African commercial abattoir from three PORCUS carcass classification groups (P, O and R) and samples from their Longissimus lumborum (LL) were obtained at 24 hours post mortem. Correlations between carcass weight, backfat depth, intramuscular fat percentage and subcutaneous fat androstenone concentrations were estimated. A sub-sample of O carcasses (n = 12) with high levels of androstenone levels were used to determine the influence of pan-frying, oven-roasting and sous vide cooking methods on meat proximate composition and fat androstenone concentration. Although O carcasses were heaviest, backfat depth increased over the carcass classifications. Intramuscular fat percentages were lowest in O carcasses, and crude protein percentages were higher in O compared with R carcasses. The correlation between percentages of backfat depth and LL intramuscular fat was insignificant.No differences were observed between classifications in androstenone concentration, which had a weak positive correlation with carcass weight. All cooked samples exceeded the established sensory threshold for androstenone (>0.45 μg/g). Cooking methods were ineffective in decreasing the potential for consumers to experience boar taint. Therefore, alternative processing methods should be considered for these carcasses. In addition, the PORCUS classification system cannot be used reliably to estimate the intramuscular fat content or androstenone concentration of a carcass. These findings are interpreted to suggest a revision of the application of the current classification system is required.