Search this question and you get a list. Almonds, greek yoghurt, dark chocolate at some suspiciously specific cocoa percentage. The lists are not wrong exactly, they are just answering a smaller question than the one being asked, and they skip the finding that actually matters.
The short answer
Protein, and the amount is what does the work rather than the source. A 2005 study in the American Journal of Clinical Nutrition put 19 adults through a sequence of diets designed to isolate one variable. First a weight-maintaining diet at 15 percent protein. Then the same energy intake at 30 percent protein, with carbohydrate held constant at 50 percent both times, so only the protein-to-fat ratio moved. Then twelve weeks at 30 percent protein with no restriction at all, eating as much as they wanted.
Satiety was markedly increased on the higher-protein version even when energy was matched. Over the free-eating phase, body weight fell by 4.9 kg and fat mass by 3.7 kg. Nobody was asked to eat less. They ate less.
What happened when protein went from 15 to 30 percent
The detail that makes this study useful is the hormonal part, because it cuts off the obvious objection. Leptin and ghrelin both moved in the direction that should have made people hungrier and driven intake back up. Leptin fell. Ghrelin rose. The body was, in effect, arguing for more food.
The reduction in intake happened anyway, and held for twelve weeks. That is a stronger result than a simple appetite questionnaire, because it means the protein effect was not merely surviving the counter-regulation, it was outrunning it.
Why protein and not the others
Protein is the macronutrient with the largest effect on the fullness signalling that follows a meal, and it is also the one most people short themselves on when eating gets disorganised. Those two facts combine badly. A day that drifts, no real breakfast, something quick at lunch, tends to be a low-protein day by accident, and low-protein days generate the evenings people then blame on willpower.
This is also why the advice is a floor rather than a rule. There is nothing to give up. There is one number to reach, and the rest of the day arranges itself around having reached it. We set out how to pick that number in how much protein you actually need.
Fibre helps, less than advertised
Fibre earns a place, with caveats. It slows gastric emptying and adds bulk, and the viscous kinds do noticeably more for fullness than the ones added to a cereal box to improve a label. The effect size is smaller than protein's and it operates mostly within and just after a meal rather than across a day. We went through which types actually change appetite in does fibre actually keep you full.
The practical version: fibre is a good passenger and a poor driver. Vegetables and legumes alongside the protein, not instead of it.
Water, honestly
Water gets promoted well beyond its evidence. Drinking a glass before a meal can modestly reduce how much is eaten at that meal, particularly in older adults, and that is roughly the size of the claim the literature supports. The idea that cravings are misread thirst signals is popular and not well supported.
Drink the water. Do not build a strategy on it.
When you eat it matters more than people expect
The 2005 study raised protein across the whole diet, not at one meal, and that detail gets dropped in most retellings. In practice the days that go wrong are rarely evenly low. They are front-loaded with almost nothing and back-loaded with everything, which means the appetite signalling that protein drives never gets going until it is too late to matter.
Spreading protein across the day is therefore doing two jobs at once. It makes the daily number achievable, because reaching a floor in one sitting is unpleasant and most people simply do not. And it puts the fullness effect where the trouble is, which is the long stretch between lunch and the evening rather than the evening itself.
This is also the honest caveat on the whole topic. Protein is a genuine lever on appetite. It is not a lever on everything, and a low-protein day is only one of several reasons an evening gets loud. Short sleep is another, stress is another, and a stretch of unusual routine is another.
Why eat-this-instead lists usually fail
The swap genre assumes the craving is for a nutrient and can therefore be redirected to a healthier carrier of that nutrient. Craving cheese, eat cottage cheese. The problem is that specific cravings track the sensory experience rather than the nutrient content, so the substitute frequently fails to end the episode and you end up eating both.
Swaps work when the substitute genuinely resembles the thing on the dimension you actually wanted, which is usually texture, salt or temperature rather than macronutrient. And they work far better as part of a day that already had enough protein in it. The protein-forward options worth keeping around are the ones that hold up on both counts.
A floor instead of a rule list
OffRamp calculates a single number: 1.6 grams of protein per kilogram of your goal weight, rounded to the nearest 5 grams. Goal weight rather than current weight, because the target is the body you are trying to hold, and stored in kilograms internally regardless of which unit you have the app set to display. If you are running a reset plan, the floor lifts by 10 percent and the rounding stays the same.
None of this is medical advice, and questions about a medication itself, including anything about starting, changing or stopping one, belong with your prescriber rather than with an app or an article. What sits safely on this side of the line is the ordinary nutrition question of whether you ate enough protein today.
There is no calorie target in the app and there never will be, which is a deliberate constraint rather than a missing feature. A floor tells you what to add. A ceiling tells you what to fear, and the research in this piece suggests the fear is doing most of the damage.
