Add up the calories of everything that goes in. Weigh the finished dish and divide total calories by total grams — that is your calories per gram. Weigh the piece on your plate and multiply.
Dividing by the number of pieces gives you the average piece. Nobody eats the average piece.
The problem with “serves six”
Type this question into a search box and every answer arrives at the same last step: add up the ingredients, then divide by the number of servings. Some pages dress it up with a calculator. The arithmetic is fine. The problem is the input.
You decided it serves six. Then you cut it, and one piece came out of the corner with all the crisp edges and another came out of the middle where the sauce pooled. Those two pieces are not the same food and they are certainly not the same weight, but the diary gets the same number for both.
A packet solves this by having the manufacturer weigh it for you. A tray bake has no manufacturer. So you take over the job, and it costs one extra step: weigh the dish when it comes out, and weigh the piece you are about to eat.
The method, in four steps
- Weigh the empty dish before you fill it and write the number down. Thirty seconds now saves you a guess later.
- Log every ingredient as it goes in. Weigh it, or use the packet weight. The oil in the pan counts. The cheese on top counts. Water and plain spices do not.
- Add up the calories, weigh the finished dish, subtract the empty weight, divide. That is the finished-weight method from our guide to calories in soups, stews and curries, and it works the same way here: total calories ÷ total grams = calories per gram.
- Weigh your piece and multiply. This is the step a pot never needs, and the step every other page on this question skips.
Why this beats a cooking-yield table. You will find charts telling you what percentage of its weight a chicken loses in the oven. You do not need one. Weighing the finished dish measures your oven, your tray and your evening directly — the yield already happened, and the scale already counted it.
A whole dish, worked through
A chicken, potato and broccoli bake in a 9×13 dish, cut into six. Not a pot — something you cut with a knife, which is where the serving problem lives.
Everything that went in
Raw ingredients weighed 2,280 g going in. The per-100 g figures above are the generic entry for each category, which is what you use when you cook from scratch rather than from a packet — check yours against the label on what you actually bought, because cheddar and chicken both vary by producer. The one number here that is not a lookup is the oil: every culinary oil is 884, and that figure can be calculated rather than looked up.
Now the part every other page skips. Say the bake comes out of the oven and the food alone — dish subtracted — weighs 1,780 g.
An illustration, and labeled as one. We have not cooked this bake for you, so that finished weight is a plausible number rather than a measurement — and so are the two piece weights further down. Yours will be different, which is exactly why the method asks you to weigh rather than to trust a page. The arithmetic that links them is the part that transfers.
The corner piece and the middle piece
Six equal pieces is the assumption. Here is what it costs when the assumption is wrong by an ordinary amount.
The same bake, three different pieces
Both of those are “one serving of the bake”. Logging them identically is a 150-calorie coin flip, and you take that flip every time you eat the leftovers.
The scale removes it in four seconds. That is the entire argument of this page.
Work out your own dish
Bring the total from your own ingredients. Nothing is stored, and no account is needed.
What you can skip weighing, and the test that proves it
Most advice hands you a list to memorize. A rule you can check yourself is better, and here it is.
| Ingredient | Typical amount in a family dish | Worst case, whole dish | Worst case, one piece of six | Verdict |
|---|---|---|---|---|
| Dried herbs and spices | 5 g | 20 kcal | 3 kcal | Skip |
| A stock cube or bouillon paste | 10 g | 40 kcal | 7 kcal | Skip |
| Garlic | 10 g | 40 kcal | 7 kcal | Skip |
| Vinegar or lemon juice | 10 g | 40 kcal | 7 kcal | Skip |
| Mustard or hot sauce | 10 g | 40 kcal | 7 kcal | Skip |
| Olive oil | 30 g | 265 kcal — it is fat, so the real figure, not a bound | 44 kcal | Never skip |
| Cheddar on top | 100 g | 403 kcal — the real figure | 67 kcal | Never skip |
In the worked example above, the oil and the cheese are 130 g out of 2,280 g that went in — under six percent of the weight — and they carry 668 of the 2,542 calories. Get those two right and everything else is detail. Why the fats do this, and how to catch them, is its own subject.
One caveat on the scale itself. Most domestic kitchen scales display whole grams, so every single reading is already rounded by up to half a gram before accuracy is even discussed. Across a dozen ingredients that is noise. It is another reason not to chase the oregano.
Raw or cooked?
Weigh raw, look up raw. That is what you are putting in, and a recipe is a list of raw things.
The trap is mixing the two. Database entries exist for both states, and a cooked entry reads higher per 100 g because the water has gone — the food did not gain anything, it just lost weight. Weigh 100 g of dry rice and log it against a cooked entry and you have understated it badly; do it the other way and you have overstated it just as badly.
With this method the question mostly answers itself: ingredients get weighed raw at the start, the dish gets weighed once at the end, and nothing in between needs a conversion. Rice and pasta are the exception worth naming, because they are the foods people most often weigh in the wrong state — weigh them dry, with the rest of the raw ingredients.
Picking the right database entry
Most of the error that survives good weighing comes from choosing the wrong line in the database.
- The same cut has several entries. Chicken thigh exists with skin, without skin, and as meat only, and they are not close to each other. Read the whole entry name, not the first three words.
- Ground meat is sold by its fat content. The percentage on the label is the entry you want. Choosing by the word “beef” alone is choosing at random within a wide range.
- Cheese and cream vary by producer. For anything with a brand on it, the panel on the packet beats a generic entry every time.
If you are wondering what makes two databases disagree about one food, mostly it comes down to which entry each one picked.
When you already ate it
No scale in the house at all? That is a different problem, and where estimating by eye actually costs you answers it directly.
The honest case, and the one nobody writes about. The tray is gone, you weighed nothing, and you still want a number that is not a fantasy.
Reconstruct forwards from what you put in, which you probably remember better than what you took out:
- List what went in the dish and be generous with the fats. This is the number you can still recover, because it came from packets.
- Count how many pieces you cut it into. That you will remember.
- Divide, and accept the average. You have lost the ability to know whether yours was the corner or the middle.
- Widen the entry honestly. Log the nominal 424, and write the band beside it — with pieces varying by a quarter that is 318 to 530. If you know you took the corner, log the top of it. Knowing your entry is 424 give or take 106 is worth more than a 424 you believe.
A reconstructed entry with a known error bar is worth far more than a blank day, and far more than a precise-looking number invented from nothing.
Where the total goes wrong after you have added it up
Baking makes a dish lighter and denser, and neither of those changes the total. Only three things do:
- Fat you pour off and throw away. If you drain the tray and bin the fat, that fat is gone. Weigh what you pour off if you want the credit for it.
- Anything added after you finished weighing. This one goes the other way — the butter on the potatoes at the table never entered the sum.
- Alcohol that boils off. Some of the wine leaves, some stays, and how much depends on the pot and the time. Do not assume all of it, and do not assume none.
Everything else — steam, browning, resting, reheating — changes the weight and the flavor and leaves the calorie total exactly where it was.
The short version
- Add up what goes in, weigh what comes out, divide to get calories per gram.
- Weigh your piece. “One of six” is a guess, and it is worth about a hundred and fifty calories.
- Write the per-100 g figure on the leftovers box and the rest of the week takes care of itself.
- Skip anything whose worst case is under 50 calories for the whole dish. Never skip the fat.
- Weigh raw, log raw, and weigh the finished dish once at the end.
Please read. This is a method for counting, not dietary advice. If you are managing a medical condition, pregnant or breastfeeding, or working to a therapeutic diet, use the figures your clinician gives you. If counting is making your relationship with food worse rather than better, stopping is a reasonable answer — in the US the National Alliance for Eating Disorders helpline is 1-866-662-1235. See our health disclaimer.
You can earn the full version instead of paying for it
If you would rather not build the ingredient list by hand, a photo of the finished piece gets you an estimate in seconds, and photo logging is on the free plan. For a dish you cooked yourself the arithmetic on this page will beat it — the camera cannot see the oil that went in an hour ago.
- Create an account and log a few meals. New accounts open with a two-day Plus trial — three days if somebody invited you — and then settle onto the free plan: no card, no end date, one meal scan a day.
- Publish an honest post, story, write-up or short video about what you found — or invite three people who go on to actually use it.
- Send us the link. A published piece unlocks the higher tier; three qualified invites unlock 90 days of the full version.
Your post has to say you got free access in exchange — that is an FTC requirement in the United States. Access is granted whether what you write is positive or critical. Full terms: promotional access.