Tim Spector Says Exercise Doesn't Cause Weight Loss. The Study Behind That Claim Is Still Being Argued Over

Tim Spector's claim that "exercise doesn't help you lose weight" and that "all the studies prove it" came from a conversational aside on Steven Bartlett's podcast The Diary of a CEO, not from a new paper. Doctissimo's write-up amplifies a soundbite that, in Spector's own words on the show, was narrower than the headline suggests: he said exercise alone, without dietary change, does little for weight loss, while crediting it for maintaining weight after diet-driven loss and for mood and cardiovascular health.

The science behind the quip is real but contested, and it has a name: the constrained total energy expenditure model, published by Duke evolutionary anthropologist Herman Pontzer and colleagues in Current Biology in 2016. Using doubly labeled water measurements in Tanzania's Hadza hunter-gatherer community, Pontzer's team found that physically active foragers burned roughly the same total daily calories as sedentary Westerners. Their explanation was metabolic compensation: the body dials down resting energy expenditure to offset calories spent moving, so a 2016 meta-analysis cited alongside the study estimated that added exercise raises total energy expenditure by only about 30 percent of what simple arithmetic would predict.

That model is far from settled consensus. A 2023 review covered by ScienceAlert argued the hunter-gatherer evidence is too thin to generalize, and pointed to randomized controlled trials showing exercise does measurably raise energy expenditure when tracked directly. Harvard evolutionary biologist Daniel Lieberman, in his 2021 book Exercised, takes the opposite position from Spector: he argues compensatory eating after exercise is modest and slow to kick in, and that exercise volumes above the standard 150 minutes a week do help sustain weight loss and prevent regain. A separate, widely cited trial found an average of about 1,000 kcal a week in compensatory eating regardless of how much people exercised, which supports partial compensation but not the zero effect implied by Spector's phrasing.

Public health guidance has not caught up to the headline version of Spector's claim either. NICE-derived clinical guidance in the UK recommends 45 to 90 minutes of daily activity specifically for weight loss and for keeping weight off, a far higher dose than the general 150-minutes-a-week target for health. That guidance still treats exercise as a legitimate weight management tool, provided the dose is high enough and diet is addressed in parallel, which is close to the maintenance-focused nuance Spector voiced on the podcast but lost in translation by the time it became a headline.

Spector's framing is also worth reading next to his business interests. He is a professor of genetic epidemiology at King's College London and co-founder of ZOE, a subscription personalized-nutrition company built on his PREDICT studies, according to a profile by Sifted. An UnHerd investigation by Deborah Cohen and Dr Margaret McCartney questioned how scientifically grounded ZOE's personalized food recommendations actually are, flagging concerns from other doctors that the app's scoring system could fuel health anxiety. None of that invalidates the Pontzer data Spector is leaning on, but a doctor whose company sells diet plans has an obvious incentive to tell the public that diet, not exercise, is the lever that matters.

What the evidence actually supports is narrower than either the Doctissimo headline or Spector's own line on the podcast: exercise without dietary change produces smaller weight loss than many people expect, because of partial metabolic compensation documented in Hadza foragers and in some lab trials, but the size of that compensation, and whether higher doses of exercise overcome it, is still being argued out between Pontzer's camp and researchers like Lieberman, in papers published years apart and measured under very different conditions.