Raw or Cooked? Which Weight to Use When You Log Food
The short answer
Weigh dry or raw, and log the entry that says dry or raw — because that is the state the packet and most reference figures are defined in. Cooking moves water in and out of food; it does not add or remove energy, so 75 g of dry pasta and the 180 g of cooked pasta it becomes are the same meal, and weighing the second while logging the first is how a 265 kcal portion turns into 650 on your screen.
Why does the weight change but the calories don't?
Heat does three things to a portion: it drives water off, it lets water in, or it lets fat run out. Only the last one actually removes energy from the food.
Rice and pasta start dry. Boiling them is mostly an exercise in absorbing water, which weighs a great deal and contains nothing. Meat starts wet. Cooking squeezes water out of the muscle, so the same protein and fat arrive in a smaller, denser parcel.
That is the whole mechanism. Everything below is arithmetic on top of it.
How much do starches gain?
Approximate yields, by weight, for the usual suspects:
White rice: 1 part dry becomes about 2.5–3 parts cooked
Dried pasta: about 2.2–2.5 parts
Rolled oats cooked with water: about 3 parts
Dried lentils and beans: about 2.5 parts
Couscous: about 2.5–3 parts
Work one through. Dry white rice runs about 360 kcal per 100 g in USDA FoodData Central; cooked white rice is about 130 kcal per 100 g, and the gap between those two numbers is almost entirely water.
So 75 g of dry rice is about 270 kcal, and lands on the plate as roughly 200 g cooked — 200 × 1.3, about 260. The two routes agree, which is the point.
Now get it wrong. Serve yourself 200 g of cooked rice, weigh it, and log "rice, dry, 200 g": 720 kcal. You have invented 450 kcal out of tap water. Go the other way — weigh 75 g dry but pick the cooked entry — and you log about 98 kcal for a 270 kcal portion.
That is a factor of nearly three in either direction, on a food most people eat several times a week. It is the single largest avoidable error in a food diary.
How much do meats lose?
Less dramatically, and in the opposite direction:
Chicken breast, roasted or pan-cooked: loses about 25% of its weight
Lean beef mince, browned: about 25–30%
Salmon fillet: about 20%
Streaky bacon: about 60–70%, which is why rashers look like they have been stolen
Raw skinless chicken breast is around 120 kcal per 100 g; cooked, about 165 kcal per 100 g. Those are not two different foods, they are the same food with a quarter of the water gone.
So 200 g raw is about 240 kcal, and comes out of the pan at roughly 150 g — 150 × 1.65, about 248. Weigh the cooked 150 g and log it as raw and you get 180 kcal, understating by about a quarter. Not catastrophic, but it compounds every day if chicken is your default protein.
Bacon is where the meat side gets as bad as the rice side. Weighing cooked rashers and logging them as raw can halve the number.
Which state should I weigh in?
Raw or dry, for anything you cook yourself. Three reasons, in order of how much they matter:
It is fixed. 75 g of dry pasta is 75 g of dry pasta in every kitchen. Cooked weight depends on how long you boiled it and how well you drained it.
It is what the packet says. Nutrition panels declare per 100 g as sold, which for a bag of rice means dry. Databases built from packaging — Open Food Facts among them — inherit that convention.
It is easier. You weigh before cooking, once, and the pan can do what it likes afterwards.
The exception is food you did not cook: a takeaway, a canteen tray, someone else's dinner. Then cooked is all you have, so use a cooked entry and say so. Canned beans are also already cooked — drained chickpeas are around 120–140 kcal per 100 g against roughly 360 for the dried ones, and that difference is water the factory added.
Consistency beats correctness here. Always raw, or always cooked, is recoverable; alternating is not, because your own history stops being comparable with itself. If you are re-reading old entries and cannot tell which convention you used, where calorie numbers come from covers how much weight to put on them.
What if I can only weigh the finished dish?
Weigh the pot, not the plate. Log every ingredient in its raw state as you put it in, total the pot's calories, then weigh the whole cooked contents and divide.
A chilli made from 500 g raw mince, 400 g rice and the rest might total 2,600 kcal and weigh 2,000 g cooked. That is 130 kcal per 100 g of finished dish, and now you can serve yourself 350 g and log about 455 without weighing anything raw again. Counting calories in batch cooking goes through the division properly, including uneven ladles.
If you have already cooked and logged nothing, use the yield factors above in reverse. 300 g of cooked pasta on the plate ÷ 2.3 ≈ 130 g dry. It is a rougher number than weighing would have given you, but it is in the right ballpark, which is the whole job. Describing it in a sentence — "about 300 g cooked pasta, so roughly 130 g dry" — is how I would log it in Day So Far, because the app takes the sentence rather than making you choose between two database rows whose names you can't tell apart.
What about food cooked in liquid that gets drained?
Divide by whether the liquid carries energy.
Water you throw away — pasta water, the water rice boiled in, the water you soaked beans in — costs you almost nothing. A little starch leaches out; it is not worth a line in the diary. Log the dry weight and move on.
Liquid you keep — stock in a soup, the sauce a stew braised in, coconut milk in a curry — is part of the dish. Its calories stay in the pot whether the meat absorbed them or not, so include them in the pot total and divide as above.
Fat you pour off is the one case where real energy leaves. Brown 500 g of 20% mince and you can pour several tablespoons of fat out of the pan, and a tablespoon of fat is about 120 kcal. That is a genuine 200–400 kcal you did not eat — but nobody drains identically twice, so treat it as a rough credit rather than a measurement, and drain the same way each time. If precision there matters to you, buying leaner mince makes the question go away.
Oil the food absorbed goes the other way and is easy to forget. Two tablespoons of oil in the roasting tin is about 240 kcal, and most of it is now on the vegetables. Weighing the vegetables raw does not capture it; logging the oil separately does.
How much does any of this actually matter?
One rice error is worth roughly 450 kcal. Repeat it four times a week and you are about 1,800 kcal a month away from where you think you are — call it a quarter of a kilo of fat at roughly 7,700 kcal per kilogram, in whichever direction the mistake points.
That is small enough to be invisible on any single day and large enough to explain a diary that says one thing while the scale says another. If yours is doing that, tracking calories but not losing weight works through the other usual suspects — but check your rice entry first. It is quicker.
None of this needs to be exact. Yield factors vary with your pan, your hob and how much you like your pasta cooked; the figures here are ordinary published averages and should be read as "about". Weighing raw and staying consistent removes the factor-of-three error, and the factor-of-three error is the one that was actually costing you something.