Is there a best diet for weight loss? Christopher Gardner answers

There is no best diet for weight loss, at least not one that trials can find. When researchers put popular diets against each other for a year or longer, the average differences were small, while people on the same diet ended up far apart. So the more useful question is which way of eating you could still be following a year from now.

Gardner got there the long way. In a study he calls A TO Z, 311 women spent a year on one of four approaches: Atkins, Ornish, Zone, or a traditional health professional's approach as the control. When the paper came out in 2007, the groups "were a few pounds different," he told Andrew Huberman. What struck him were "the within-group differences, which were massive in every one of the groups." Some women lost a lot of weight; others on the same approach gained. "Oh my God," he remembers thinking, "the difference within the diets is way cooler than the average difference between the diets" (Huberman Lab, from 56:41).

So he set out to predict who would do better on which diet. That became DIETFITS, the larger trial below, and the prediction failed. By 2025, when Huberman opened by asking whether there is a best diet, Gardner's answer was blunt: "So there isn't one best diet, and I don't think we need different diets" (from 02:32). He meant eating well in general, and the weight-loss trials below point the same way.

What happened when diets went head to head?

1. Low-carb against low-fat, for a full year

DIETFITS randomized 609 adults in California, aged 18 to 50, with a body mass index of 28 to 40 and without diabetes, to a healthy low-fat or a healthy low-carb plan for 12 months, taught in 22 small-group classes (Gardner et al., 2018). Both groups cut way back on fat or carbs for two months, then added some back until they reached the lowest level they believed they could keep "indefinitely." Neither group had a calorie target. Both were told the same things about quality: lots of vegetables, little added sugar or refined flour, whole foods cooked at home.

After a year the low-fat group had lost 5.3 kg on average and the low-carb group 6.0 kg. The 0.7 kg gap wasn't statistically significant. Within each group, weight change ran from about 30 kg lost to about 10 kg gained. The trial also tested whether a gene pattern or how much insulin people released after sugar could predict who did better on which diet. Neither could.

Gardner's own summary: "you could do either one. If you do them in a healthy way, it would be okay" (Huberman Lab). The trial's population was narrow: adults under 50 in the San Francisco Bay Area, with a year of classes that few diets come with.

2. More than a hundred trials, the same answer

A 2020 review in The BMJ combined 121 randomized trials with 21,942 adults covering low-carb, low-fat and moderate patterns and 14 named diets, Atkins and Zone among them (Ge et al., 2020). At six months, almost every diet beat the participants' usual eating, but the differences between diets were "generally trivial to small." By 12 months the weight loss had shrunk for nearly all of them.

The POUNDS Lost trial ran longer. It gave 811 adults in Boston and Baton Rouge one of four mixes of fat, protein and carbohydrate for two years (Sacks et al., 2009). The groups' weight changes differed by less than half a kilo, and at two years no mix had pulled significantly ahead. What did track weight loss was attendance: people who came to more counseling sessions lost more, whichever diet they had. The authors read that link as engagement, not proof.

Obesity Canada's nutrition guideline, written for doctors and dietitians, reads the same literature this way: "no single best nutrition intervention has been shown to sustain weight loss long-term, and literature continues to support the importance of long-term adherence, regardless of the intervention" (Brown et al., 2022).

3. Fasting and intuitive eating, measured the same way

Intermittent fasting has been tested the same way. A 2026 Cochrane review of 22 trials with 1,995 adults found that, compared with regular dietary advice, intermittent fasting "may result in little to no difference in weight loss" over up to a year, with low certainty in the evidence (Garegnani et al., 2026). A 2025 review of 99 trials found fasting plans and ordinary calorie cutting reduced weight by similar amounts, apart from a small edge for alternate-day fasting in shorter trials, and in the trials lasting six months or more none of the fasting plans beat steady calorie cutting (Semnani-Azad et al., 2025).

Gardner agrees. "The data are not very compelling, in my opinion, for intermittent fasting," he told CNN's Sanjay Gupta, partly because there are so many versions (Chasing Life, 18:20). He grants the appeal: the rule is simple, and "it's not time to eat" is easy to follow.

Mindful and intuitive eating, which ask you to eat by hunger and fullness instead of a plan, landed in the same place: in a review of 10 short randomized trials, weight change was no different from conventional diet programs (Fuentes Artiles et al., 2019).

Why does weight loss on most diets fade?

4. Most diets work for months, then lose ground

The trials share a shape: weight comes off in the first six months and some of it returns after that. In the BMJ review, the effects of all the named diets were smaller at 12 months than at six. In POUNDS Lost, people "began to regain weight after 12 months" on every diet.

Sandra Aamodt, a neuroscientist who spent three decades on and off diets, put the frustration in one line in her TED talk: "If diets worked, we'd all be thin already" (TED, 08:56). It's a joke, and the trials say something narrower: diets do work for a while, about equally, and the hard part is the months after.

You may have read that 95% of diets fail. No traceable study supports that figure. The honest version: many people who lose weight by dieting regain some or all of it (Montani et al., 2015).

5. Strict rules meet ordinary days

Traci Mann, a psychologist who runs the Health and Eating Lab at the University of Minnesota and the author of Secrets from the Eating Lab (HarperCollins), started out looking for ways to stop dieters from overeating. "Nearly everything I studied (e.g., stress, distraction and others) caused dieters to lose control of their eating," she told NPR, and she says willpower is "in many ways a mythical quality and certainly not something that can be relied upon for weight loss" (NPR). These are her conclusions; the interview doesn't show the studies.

On Sigma Nutrition Radio, the hosts Danny Lennon and Alan Flanagan discuss a study of eight everyday eating behaviors. People who changed more of them did best over three years, and rigid control of food intake was one of the two behaviors not linked with keeping weight off (Sigma Nutrition Radio, 21:11). The study is named only from memory and isn't one we could check, so treat this as their reading.

Gardner designed DIETFITS around exactly this problem. "If it's only a diet you go on and then a diet you will go off when you're done. That won't help. It has to be something that you can maintain," he told Gupta (06:25).

How do you choose a way of eating you can keep?

The trials can't pick a diet for you. They do suggest what to look at.

6. Start from the way you already eat

DIETFITS didn't ask anyone to adopt a new cuisine. People cut back on one thing and then added it back until they found a level they could live with. Obesity Canada's guideline asks for the same starting point: an approach that fits your "values, preferences and treatment goals," and that is "culturally acceptable and affordable for long-term adherence."

If your weekday lunches are already a sandwich and fruit, a change that keeps the sandwich is more likely to survive March than a plan that bans bread.

7. Notice whether you're hungry on it

Gardner suspects fullness explains much of the spread in his trials. "I haven't quite figured out how to do this study yet, but I think it's hugely satiety or satiation," he told Gupta (08:51). He's clear that it hasn't been tested. In DIETFITS, the instruction was blunt: "we don't want you to be hungry. We want you to eat till you're satiated" (06:25). Our guide to always feeling hungry goes further.

8. Plan for the week that goes wrong

Every plan meets a birthday, a late shift or a bad night. A rule that one slip breaks can take the whole plan down with it, while a rule with room for the slip lets the next day go as planned. Our article on cheat days and all-or-nothing eating covers that moment, and weight loss plateaus covers the weeks when nothing seems to move.

That last question is yours to answer. Some people track how their evenings feel; some want a number on the scale or a waistband. Either is a fair thing to watch. The trials can't tell you which plan gets you there; they do say the differences between plans were small.

Back to the women in A TO Z

Gardner's surprise in A TO Z was how little the label on the diet book mattered. DIETFITS and the review of 121 trials found the same. Why people on the same plan ended up so far apart is still open, and his best guess is fullness. What the trials agree on is that keeping it up mattered more than the label. As he told Huberman, "Nutrition scientists don't really disagree. We're almost boringly more in agreement than most people think." For more on what lasts, start with our guide to how to lose weight sustainably.

To build that around your own days, try the 1h Protocol to Lose Weight Sustainably. It's practical education about everyday eating and movement habits, with no meal plan, calorie target or weight set for you. The first chapter is free: a few questions find the pattern your eating follows and give you one thing to do tomorrow. The full program takes about an hour. It walks one day with you, from the first meal to bed, at your own times, adds a plan for the "just one" moment, and lets you write your own goal, a number included if you want one. You leave with your chosen changes on one page, a card for your lock screen, a fourteen-day sheet and days 7 and 14 in your calendar, and short check-ins by email follow on days 1, 7, 14 and 30, each answered with one tap. It isn't for anyone pregnant or breastfeeding, or with an eating disorder now or in the past. Start free

Frequently asked questions

What is the best diet for weight loss?

No diet has won in trials. In a review of 121 trials, differences between named diets were "generally trivial to small" and most of the effect had faded by 12 months (Ge et al., 2020). In DIETFITS, healthy low-fat and healthy low-carb plans ended the year 0.7 kg apart (Gardner et al., 2018). The way of eating you can keep is the best candidate you have.

Why do diets fail?

Most don't fail at first: in trials, people lose weight for about six months and then some of it returns (Sacks et al., 2009). Strict plans meet stress, distraction and ordinary weeks, and the psychologist Traci Mann says nearly everything she studied made dieters lose control of their eating (NPR). That's why Obesity Canada's guideline puts long-term adherence ahead of the type of diet (Brown et al., 2022).

Does intermittent fasting work better than other diets?

Not in the reviews so far. Compared with regular dietary advice, it "may result in little to no difference in weight loss" (Cochrane, 2026). In a year-long trial of 100 adults in Chicago, alternate-day fasting "did not produce superior adherence, weight loss, weight maintenance" compared with daily calorie restriction, and more people dropped out of it, 38% against 29% (Trepanowski et al., 2017). Eating only between noon and 8 pm did no better than three meals a day in a 12-week trial of 116 adults (Lowe et al., 2020). If fasting suits your days, it's one option among several.

Is yo-yo dieting dangerous?

Researchers haven't settled it. A 1994 US task force found that "the majority of studies do not support an adverse effect of weight cycling on metabolism," and that studies linking weight swings with illness mostly couldn't tell intentional loss from unintentional loss (National Task Force, 1994). A 2015 review called the question "the source of considerable controversy," with findings that "have not been consistent" (Montani et al., 2015). So "yo-yo dieting wrecks your metabolism" isn't supported, and neither is the opposite claim that weight cycling is proven harmless.

Does intuitive eating help with weight loss?

About as much as a conventional diet, in short trials. A review of 10 randomized trials found no difference in weight change between mindful or intuitive eating programs and conventional diet programs (Fuentes Artiles et al., 2019). Another review of 13 trials found most showed no clear difference in how much people ate, and rated the studies weak (Grider et al., 2021). Sandra Aamodt, who learned to eat this way over about a year, says it "probably won't make you lose weight unless you often eat when you're not hungry" (TED, 08:56).

Sources

This article draws on Christopher Gardner's interviews on Huberman Lab and CNN's Chasing Life, Sandra Aamodt's TED talk, Traci Mann's interview with NPR, an episode of Sigma Nutrition Radio, and the trials, reviews and guideline listed below.

• How Different Diets Impact Your Health, with Dr. Christopher Gardner (podcast episode, Huberman Lab, 2025)
• Do Diets Really Work?, with Christopher Gardner (podcast episode, CNN Chasing Life, 2024)
• Gardner et al, Effect of low-fat vs low-carbohydrate diet on 12-month weight loss: the DIETFITS randomized clinical trial (randomized trial, 609 adults aged 18 to 50 without diabetes, JAMA, 2018)
• Ge et al, Comparison of dietary macronutrient patterns of 14 popular named dietary programmes (network meta-analysis, 121 trials, 21,942 adults, BMJ, 2020)
• Sacks et al, Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates (randomized trial, 811 adults aged 30 to 70, two years, NEJM, 2009)
• Brown et al, Medical nutrition therapy in obesity management (Canadian Adult Obesity Clinical Practice Guidelines, chapter 8, version 2, 2022)
• Garegnani et al, Intermittent fasting for adults with overweight or obesity (Cochrane systematic review, 22 trials, 1,995 adults, abstract and key messages, 2026)
• Semnani-Azad et al, Intermittent fasting strategies and their effects on body weight and other cardiometabolic risk factors (network meta-analysis, 99 trials, BMJ, 2025)
• Trepanowski et al, Effect of alternate-day fasting on weight loss, weight maintenance, and cardioprotection (randomized trial, 100 adults, one year, 2017)
• Lowe et al, Effects of time-restricted eating on weight loss: the TREAT randomized clinical trial (randomized trial, 116 adults, 12 weeks, 2020)
• Fuentes Artiles et al, Mindful eating and common diet programs lower body weight similarly (systematic review and meta-analysis, 10 trials, abstract only, 2019)
• Grider et al, The influence of mindful eating and/or intuitive eating approaches on dietary intake (systematic review, 13 trials, abstract only, 2021)
• Sandra Aamodt, Why dieting doesn't usually work (TED talk, 2013)
• In 'Eating Lab,' a psychologist spills secrets on why diets fail (interview with Traci Mann, NPR, 2015)
• Traci Mann, Secrets from the Eating Lab (book, publisher page, 2015)
• Do Diets Even Work in the Long-term? (podcast episode, Sigma Nutrition Radio, episode 352, 2020)
• National Task Force on the Prevention and Treatment of Obesity, Weight cycling (consensus review, abstract only, JAMA, 1994)
• Montani, Schutz & Dulloo, Dieting and weight cycling as risk factors for cardiometabolic diseases (narrative review, Obesity Reviews, 2015)