Woman measuring her waist, diet, weight loss, fat, torso, stomach.

Pro-active ... exercise to lose fat and gain muscle.

It's a message that's got louder over the last few years: exercise doesn't work for weight loss. There was the 2009 Time magazine cover story The Myth about Exercise featuring a photo of a woman pounding a treadmill, her eyes trained on a cream-topped cupcake. It symbolised the story's thrust: research suggesting exercise won't help weight loss because it makes us eat more to make up for the kilojoules we burn. It's a message repeated recently in the new book Big Fat Lies in which author David Gillespie - a lawyer – describes exercise for weight loss as 'pointless'. But is it really true?

No, says David Driscoll - not a lawyer, but an exercise physiologist and sports dietitian - who thinks that exercise deniers, like climate change deniers, don't tell the whole story – they cherry-pick the evidence that suits their message instead.

"The level of physical activity we'd have done a century ago just to get through the day would seem like an extreme sport compared to what most of us do now." 

It's not hard to find research showing exercise does shift weight, especially when it's combined with a leaner diet, says Driscoll, a member of Sports Dietitians Australia. But when studies do find little or no weight loss benefit from exercise, this may have more to do with behaviour than biology.

"Exercise can make us eat more - what's not clear is whether this is because our bodies are demanding the extra fuel or because we think we deserve it," he says. "In studies where people do consume more kilojoules after exercise no one's asked them if it's because of real hunger or because they feel they need a reward for working out. We need to establish this - it's well-known that eating isn't always related to hunger."

If you're in the 'I need a reward' camp – and you wouldn't be alone – knowing more about how many kilojoules you've burned can help.  If you're doing a cardio class at the gym at a moderate intensity, for instance, you're probably burning eight to 12 kilojoules a minute. Over a period of 45 minutes that adds up to between 360 to 540 kilojoules. Based on that, says Driscoll, it makes sense to settle for a small reward – two squares of chocolate not the whole bar.

"Timing an exercise class before a meal is another strategy – you'll be eating anyway and less likely to overcompensate," he says. "Another problem is that some people do a 45-minute gym class and think it means they can do nothing for the rest of the day, but 45 minutes isn't enough. You need to think 'how can I add to this?'"

One way of putting this 45 minutes into perspective is that the level of physical activity we'd have done a century ago just to get through the day would seem like an extreme sport compared to what most of us do now, he says.

When research finds little weight loss from exercise, Driscoll believes it's also important to look at how much fat is shed, not just weight. In some studies people may not lose much weight but they may lose fat and gain muscle – and muscle weighs more than fat.

As for a shining example that diet and exercise do work, there's the US National Weight Control Registry, a project tracking the progress of 10,000 people who've lost an average of 30 kilos and kept it off for over five years – and guess what? Ninety eight per cent of these successful losers report that they changed their diet in some way to lose weight – and 94 per cent increased their physical activity.

So is the exercise-is-useless message a dangerous one?

"It's irresponsible, but it's also a message that some people want to hear and it appeals to people who don't like exercise. But I don't think it will make people who are already exercising stop," says Driscoll. "And in the unlikely event that exercise turned out not to work and actually caused weight gain, the health benefits of physical activity are so great they'd offset the problem of a little extra weight."

Has exercise helped you trim down?