People with diabetes are at high risk of cardiovascular morbidity and mortality, especially if they have already developed vascular problems. For patients who are apparently free of vascular complications, risk tables are often used to assess the risk of cardiovascular events in the following years, and to decide on treatment with statins or anti-platelet therapy. These risk prediction tables include estimates of traditional cardiovascular risk factors and are based on populations, some of which only contained a very small number of people with diabetes. Multiple problems can be identified with these tables, and many seriously underestimate cardiovascular risk in people with diabetes. Possible ways of addressing this include using risk estimation tools based solely on diabetic populations, adding in additional traditional variables such as triglycerides or left ventricular hypertrophy, including novel cardiovascular risk factors, or intervening at a lower level of estimated risk in people with diabetes compared with non-diabetic subjects. Alternatively, estimates of individual risk could be abandoned and all people with diabetes could be treated with statins and other effective agents.