The promise attached to intermittent fasting is not that you will eat less. It is that you will lose weight faster for the same effort — that the clock does something the calorie arithmetic does not, by flipping a metabolic switch somewhere around the sixteenth hour.
Three randomised trials have now tested three different fasting schedules against ordinary daily calorie restriction, over a year each. They produced three different answers, which sounds like confusion and is not. Line them up and they agree on something more useful than the headline.
Start with the most popular version, because it is the one that fails
Sixteen-eight — eat between noon and eight, drink water the rest of the time — is the version most people mean. In 2020 Ethan Weiss’s team at UCSF ran it properly: 116 adults with overweight or obesity, randomised either to 16:8 eating ad libitum inside the window, or to three structured meals a day.
The fasting group lost 1.17% of body weight. The control group lost 0.75%. The difference was not statistically significant.
1.17% vs 0.75%
weight lost over twelve weeks on 16:8 versus three meals a day — a gap the trial could not distinguish from noise
There was one significant difference, and it went the wrong way. The time-restricted group lost significantly more lean mass than the controls, with no significant difference in fat loss between them. Whatever came off was disproportionately not fat.
What these trials can and cannot tell you
They are randomised comparisons in adults with overweight or obesity, run for twelve weeks to a year. They answer one question well: does this schedule beat ordinary calorie restriction? They do not tell you what a schedule will do for you personally, and they are not long enough to settle what happens over a decade.
They also cannot tell you a protocol is useless. A trial that finds no difference from calorie restriction has found something that works as well as calorie restriction, which is not nothing — it is just not the claim being made for it.
The version that did slightly worse
Alternate-day fasting is the strictest common protocol: roughly a quarter of your energy needs on fast days, a hundred and twenty-five per cent on the days between. Krista Varady’s group ran it against daily restriction for a full year, in a hundred adults, with a no-intervention control arm.
It produced no advantage in weight lost, in weight kept off, or in the cardiovascular markers. What it did produce was more dropout — people found it harder to stay with — and one finding nobody was looking for: by month twelve, LDL cholesterol was significantly higher in the fasting group than in the daily-restriction group, by 11.5 mg/dL.
And the version that did better
This is where a debunk would stop, so it is worth being careful. In April 2025 a team at the University of Colorado published the trial that goes the other way.
A hundred and sixty-five adults, twelve months, 4:3 fasting — three fast days a week — against daily calorie restriction, and crucially both groups were prescribed the same weekly calorie deficit, 34.3%. The fasting group lost 7.7 kg. The daily-restriction group lost 4.8 kg.
−7.7 kg vs −4.8 kg
at twelve months on 4:3 fasting versus daily restriction, from an identical prescribed weekly deficit — and adherence was higher in the fasting group at both six and twelve months
So intermittent fasting can beat daily calorie counting. That result is real and it is recent and it belongs in any honest account of this.
What the three trials agree on
Put them in order of how hard each schedule pushes on how much you actually eat, and the results stop looking contradictory.
16:8 does not ask you to eat less at all. You eat what you like inside eight hours. It produced no meaningful weight loss, which is what a protocol with no deficit in it should produce.
Alternate-day fasting asks for a large deficit in an arrangement most people struggled to keep. It produced no advantage and more dropout.
4:3 asks for the same deficit as daily restriction in a shape people found easier to hold — and that is the arm where adherence was higher and the weight difference appeared.
The schedule is not doing metabolic work the arithmetic cannot do. It is doing adherence work. Which is a smaller claim than the one on the packaging, and a more useful one, because adherence is a thing you can actually predict about yourself.
- Check whether the protocol contains a deficit. If a rule only tells you when to eat and never how much, the trials say to expect very little. That single test disqualifies the most popular version.
- Pick the shape you can keep, not the one that sounds most severe. The trial where fasting won is the one where people stayed with it. The trial where it lost is the one they dropped out of.
- Protect protein and resistance work if you fast. The lean-mass finding is the most actionable thing in this literature and the least repeated: weight is not the same as fat, and one of these trials shows the difference.
- Give it a year, not a fortnight. Every trial here ran twelve weeks to twelve months. Two-week verdicts are measuring water and enthusiasm.
- Treat any protocol you cannot describe a deficit for as a schedule, not a diet. It may still suit you. It is not the thing that makes the weight move.
The first of those is the only one that needs doing today, and it takes about a minute.
Try this
Take whatever eating rule you are considering — or already following — and answer one question about it: does it change how much you eat, or only when?
Most people find they cannot answer immediately, which is itself the answer. A protocol whose deficit you cannot state is one you are hoping will work by mechanism rather than by arithmetic, and that is precisely the hope these three trials were built to test.
The parts of this that are genuinely hard
The LDL result from the alternate-day trial has not been widely replicated and should not be read as a general warning about fasting. It should be read as a reminder that a protocol popular enough to have a name is not thereby known to be safe over years, because that is not what any of these trials measured.
There is also a group for whom this framing is wrong rather than merely unhelpful. If you have a history of disordered eating, a rule that makes not-eating a daily achievement is not a neutral tool, and the research on weight-loss protocols does not address that risk at all. If you take medication that has to be taken with food, or you are pregnant, or you manage diabetes, the schedule is a clinical question rather than a lifestyle one: talk to your clinician before you change when you eat, and do not stop, move or skip a dose to fit a protocol you read about.
And none of this touches the largest finding in the field, which is that most people regain. Every trial here is a comparison between two ways of losing; none of them solved keeping.
The bottom line
Intermittent fasting is not a shortcut past the arithmetic. It is a container for the arithmetic, and the trials rank almost exactly in the order of how well each container helped people hold the deficit inside it.
Which means the honest question is not whether fasting works. It is whether this particular schedule will get you to eat less for a year — and you know more about that than any trial does.
Sources (3)
Every factual claim in this article comes from one of these sources.
- Effects of time-restricted eating on weight loss and other metabolic parameters in women and men with overweight and obesity: the TREAT randomized clinical trial
- Effect of alternate-day fasting on weight loss, weight maintenance, and cardioprotection among metabolically healthy obese adults: a randomized clinical trial
- The effect of 4:3 intermittent fasting on weight loss at 12 months: a randomized clinical trial
Keep exploring
Continue reading
Deeper into what you just read.
Explore more
Adjacent ground, if this was useful.
One thing worth knowing
A surprising finding. The evidence behind it. And what it means for your life. BetterYou+ sends one useful idea at a time—nothing to keep up with.