Physical activity and exercise are frequently maligned in the popular press and in scientific journals in part because there is so much confusion. Whilst structured bouts of exercise are usually only a small part of total energy expenditure, overall physical activity is a demonstrably important component of energy expenditure and energy balance. Furthermore, specific forms of structured exercise have very potent effects on various physiological pathways and processes – irrespective of their (relatively) modest impact on energy expenditure and balance. The most powerful demonstration of this effect is in studies that have deliberately created an energy surplus alongside the imposition of regular exercise (e.g., overfeeding studies where there is deliberate weight gain alongside regular exercise). Broadly, these studies support the notion that it is indeed possible to ‘outrun a bad diet’ at least in the short term. These effects have even been observed in the tissues where excess energy is being stored (i.e., adipose). These findings are likely to be explained in part by the transient effects elicited by each bout of exercise – including the way in which exercise acutely affects the handling of ingested nutrients. The precise details and mechanisms have yet to be determined but there are several likely candidates, although it is quite probable that different forms of structured exercise exert their effects in subtly different ways. Interestingly, exercise might have a less powerful and/or distinct physiological role when introduced alongside an energy deficit from calorie restriction. This notwithstanding, the evidence is strong enough that public-facing messages should underscore the importance of the unique physiological and health benefits of certain forms of exercise irrespective of energy balance. This message needs to emphasise that the benefits of specific forms of exercise for various health outcomes will not automatically mean that these forms of exercise are appropriate for energy balance. At present, current physical activity public health guidelines are woefully unsuitable for recommending the likely amount of activity required for energy balance and weight maintenance. Future research will need to characterize the specific forms of exercise and physical activity that elicit biologically-meaningful changes with and without a major impact on energy balance. This needs to include studies that examine the way in which physical activity interacts with and affects the responses to the foods that we consume. We should also reflect on the fact that these interactions are likely to be bidirectional – for example – food intake appears to have the potential to influence free-living physical activity energy expenditure. More broadly, physiologists need to use their expertise to make a greater contribution to the development of future physical activity guidance if we are to ensure that physical activity plays a more successful role in public health.