Dietary carbohydrate restriction in the treatment of diabetes and metabolic syndrome is based on an underlying principle of control of insulin secretion and the theory that insulin resistance is a response to chronic hyperglycemia and hyperinsulinemia. As such, the theory is intuitive and has substantial experimental support. It has generally been opposed by health agencies because of concern that carbohydrate will be replaced by fat, particularly saturated fat, thereby increasing the risk of cardiovascular disease as dictated by the so-called diet-heart hypothesis. Here we summarize recent data showing that, in fact, substitution of fat for carbohydrate generally improves cardiovascular risk factors. Removing the barrier of concern about dietary fat makes carbohydrate restriction a reasonable, if not the preferred method for treating type 2 diabetes and metabolic syndrome. We emphasize the ability of low carbohydrate diets to improve glycemic control, hemoglobin A1C and to reduce medication. We review evidence that such diets are effective even in the absence of weight loss.