Aims A very limited number of prospective studies have reported conflicting data on the relation between heart rate and diabetes risk. Our aim therefore was to determine in a large, national, population-based cohort if heart rate predicts the development of diabetes. Methods The Australian Diabetes Obesity and Lifestyle study followed up 6537 people over 5 years. Baseline measurements included questionnaires, anthropometrics and blood and urine collection. Heart rate was recorded in beats per min (Dinamap). An oral glucose tolerance test was performed at baseline and follow-up, and diabetes was defined using World Health Organization criteria. Results A total of 5817 participants were eligible for analysis, 221 of whom developed diabetes. Compared with participants with a heart rate < 60 b min–1, those with a heart rate ≥ 80 b min–1 were more likely to develop diabetes (odds ratio 1.89, 95% CI 1.07–3.35) over 5 years, independent of traditional risk factors. This relationship was highly significant, particularly in non-obese men (odds ratio 5.61, 95% CI 1.75–17.98), but not in their obese counterparts or in women. Conclusions Resting heart rate is associated with an increased risk of diabetes over a 5-year period, particularly among non-obese men. This suggests that sympathetic overactivity may be a contributing factor to the development of diabetes, and that resting heart rate may be useful in predicting risk of Type 2 diabetes in non-obese men.