Less than the internet suggests, and more than nothing. The best randomised evidence says that squeezing food into a shorter window does not, by itself, produce extra weight loss. But observational research keeps finding that people who eat their main meal late lose more slowly than people who eat it early. Both findings are solid, and the reason they point in different directions is the genuinely useful part of this question.
The short answer
Clock time is not doing much work on its own. When researchers hold food constant and change only the hours it is eaten in, the weight difference mostly disappears. When they observe people eating freely, later eating does track with slower progress, because evening is when appetite runs highest and eating is least deliberate. So timing matters as a behavioural signal, not as a metabolic switch. If you eat late and it goes well, nothing about the hour is working against you. If you eat late and it tends to run long, the hour is worth paying attention to for reasons that have nothing to do with metabolism.
The trial that found nothing
The strongest single piece of evidence here is a 12-month randomised trial published in the New England Journal of Medicine in 2022. Researchers assigned 139 adults with obesity to either time-restricted eating, meaning all food between 8am and 4pm, or the same daily food eaten without any window restriction. Everyone was on the same reduced intake. After a full year, the time-restricted group had lost about 8 kilograms and the unrestricted group about 6.3 kilograms, a difference that did not reach statistical significance. Waist circumference, body fat, and blood pressure showed no meaningful separation either. The trial is unusually clean on this question because it controlled the one thing most timing studies cannot: how much food was actually eaten.
The study that found something
Now the other side. A 2013 study in the International Journal of Obesity followed 420 people through a 20-week weight loss programme in Spain, where lunch is the largest meal of the day. Participants who ate lunch after 3pm lost significantly less weight than those who ate before 3pm, and they lost it more slowly, despite no difference in how much they reported eating or how much they moved. Garaulet and colleagues also noted the late eaters slept less and skipped breakfast more often, which starts to hint at the explanation.
One study changed the clock and held the food constant. The other watched the clock and let the food do whatever it did. That is the whole difference.
Why both results can be true
Late eating is rarely just late eating. It travels with a cluster of other things: shorter sleep, skipped or minimal breakfast, less structured meals, and more grazing after dinner. When a study observes people eating freely, it picks up that whole cluster and labels it timing. When a trial controls intake and assigns windows at random, the cluster gets broken up, and the clock on its own stops predicting much.
This is worth sitting with, because it changes what you would actually do about it. If late eating were metabolically harmful, the answer would be to move your meals earlier and nothing else. Since the evidence points at the cluster instead, moving dinner from 9pm to 7pm while everything else stays the same is unlikely to do much. Fixing the sleep, or getting enough protein in before evening, plausibly does.
Is eating late at night different
Somewhat, and for one specific reason. Appetite has a real circadian rhythm, and it peaks in the evening rather than the morning. Controlled research where participants were kept on a fixed schedule, with food and activity held constant, still found hunger ratings highest around 8pm and lowest around 8am. So evening eating is not being punished by your metabolism. It is happening at the hour when the drive to eat is naturally strongest and the least amount of deliberate thought is going into it, which is a behavioural problem rather than a biochemical one. We go deeper into that pattern in why you crave sugar at night.
What to watch instead of the clock
Two things predict how a day goes far better than what time you ate. The first is whether the day's protein total landed where it needed to, because underfed afternoons reliably become difficult evenings. The second is whether the weekly weight trend is moving where you want it, which is a much steadier signal than any single day's timing. Both of those are things you can check. Neither requires setting an eating window or watching a clock.
Building OffRamp, this is exactly why there is no meal-time field anywhere in the app and no fasting timer. It was a deliberate call, made after reading this literature: adding a timestamp to every log would imply the timestamp mattered, and the controlled evidence does not support that. Logging is a daily total against your protein floor, and it does not care whether the food arrived at noon or at nine.
How OffRamp helps
OffRamp tracks the two things the research says actually predict how maintenance goes: whether you hit your protein floor, and where your weight trend is heading. Regain Radar smooths the day-to-day noise into a line you can read, so a late dinner never looks like a setback and a genuine upward drift never gets lost in it.
