Many incident cases of type 2 diabetes do not fulfil the metabolic syndrome, which accordingly has been questioned both as a research and clinical tool. The aim of this study was to determine differences in risk factors for type 2 diabetes between groups with high or low metabolic score. The study population were 26,093 men and women attending the Tromsø Study in 1994, followed through 2005, and who did not have diabetes when entering the study. A total of 492 incident cases of type 2 diabetes were registered. A metabolic score was defined according to a modified version of the National Cholesterol Education Program Adult Treatment Panel III. For those fulfilling ≥ 3 metabolic score criteria, increasing age, body mass index (BMI), triglycerides and a family history of diabetes were independent predictors. Age, BMI, and triglycerides predicted type 2 diabetes more strongly in subjects with low metabolic score, whereas high HDL cholesterol was not protective in this low risk group. The risk associated with a positive family history was unaffected by level of metabolic score. In addition smoking, low education and in men also physical inactivity were independent risk factors only in those with low metabolic score. Adding these non-metabolic risk factors increased correct classification from an ROC area of 77.2 to 87.1% (P value < 0.0001). One half of the incident cases of type 2 diabetes were missed by using high metabolic score for risk prediction.