Estimates of dressing percentage are used to predict carcase weight from live weight measurements and are affected by nutritional factors and the breed, weight and fatness of the animal. Data from the Sheep CRC's Information Nucleus Flocks for the 2007 lambing season were used to determine the effect of breed and environment on dressing percentage and aspects of hot carcase weight. Four sire groups, Merino, Border Leicester, Poll Dorset and White Suffolk, were included in the data set. Only progeny from Merino ewes (N = 1270; seven sites) were included in the data set. The dressing percentage of lambs from terminal sires (46%) was similar to that of the progeny of Border Leicester sires (45%) and higher (P < 0.01) than that of the progeny of Merino sires (43%). The live weight and fat depth at the GR site (110 mm lateral to the vertebra of the 12th rib) of the progeny was not closely related to dressing percentage. Eye muscle area explained 56% of the variation in dressing percentage based on progeny group means. Variation in dressing percentage across the seven sites was not strongly associated with feed type (pasture vs grain), GR fat depth or weight. It was concluded that the existing method of estimating carcase weight from live weight, fat score and nutrition is not accurate enough for commercial transactions involving the sale of sheep for slaughter or for breeding programs aimed at improving meat production. The sale of sheep on a hot carcase weight basis and progeny testing based on direct carcase measurements are likely to improve breeding and management strategies aimed at producing high-quality sheep meat.