How to Estimate Portion Sizes Without a Scale

The short answer

Four hand measures cover most of a plate: a palm for protein, a fist for vegetables or cooked grains, a cupped hand for dry carbohydrates, and a thumb for fat. They are rough, they are personal to your own hand, and they are far better than no reference at all.

The energies below are calculated with the same factors used across this site, 4 kcal per gram for protein and carbohydrate and 9 for fat, so they can be checked against any label.

Why hand measures work at all

Hand size scales with body size, so a larger person both needs more food and carries a larger reference. That self-scaling is why the method survives in dietetic practice despite being obviously imprecise.

The second reason is that errors shrink when averaged. Estimating every meal within 20 percent still lands the week's total close to the truth, because the overestimates and underestimates cancel rather than compound. A scale is more accurate on Tuesday; the hand is accurate enough by Sunday.

The third reason is adherence. A method you actually use beats a method you abandon, and nobody carries a scale to a restaurant. Estimation is the only approach that survives real life, which is why it belongs in the plan rather than being the fallback.

The four measures in more detail

A palm is the width and thickness of your own palm, without fingers. For chicken, fish or lean beef that is about 95 g cooked and roughly 30 g of protein, which covers a third of a 135 g daily target. Two palms at a meal is a large portion and is worth logging as such.

A fist is a serving of vegetables or of cooked grains, and it is where carbohydrate portions usually go wrong, because restaurant portions are one and a half to two fists. A cupped hand is for dry carbohydrate, rice, oats or cereal before cooking, and it is a poor measure of cooked grains because they swell.

Plate rules and the size of the plate

Half the plate vegetables or salad, a quarter protein, a quarter starch is a robust default for a mixed meal, and it works because it fixes proportions rather than grams. It needs no arithmetic and it survives a buffet.

Plate size does the rest of the work without being noticed. The same serving looks smaller on a large plate, and the reflex is to add more; the same serving on a smaller plate is read as generous. Using a smaller plate and a taller, narrower glass is one of the few changes that reliably reduces intake without any conscious effort.

The pack size trap

Servings per pack is the line to read first. A 500 ml bottle of juice that carries two and a half servings is not one drink, and a bag of nuts with four servings is not a snack. The front of the pack is written to look like a single portion; the panel behind it rarely agrees.

The practical rule: divide by servings before reading the energy. 220 kcal per serving with three servings in the pack is a 660 kcal pack, and that is the number worth knowing when the pack is going to be empty by the end of the evening.

Where the eye fails completely

Energy density decides whether estimation works. Vegetables, fruit, potatoes, rice, yoghurt and lean meat are forgiving, because a large error in weight is a small error in calories. Oils, butter, nut butters, cheese, nuts and chocolate are not: a thumb of oil and a generous pour differ by 90 kcal in the same pan.

These are the foods worth keeping measured, at least while a target matters. Two tablespoons of oil instead of one is 126 extra calories a day, which on its own is more than half of a 500 kcal deficit at 75 kg. Nothing else in the kitchen has that leverage.

Calibrate for a fortnight

Weigh the foods you eat most often for two weeks and note what your hand measures actually weigh. Most people find their palm is 20 to 40 g heavier than the reference and their thumb of oil is double. The correction sticks, and it costs two weeks.

After that, weigh occasionally rather than habitually: a new food, a new recipe, or a fortnight of logging when a plan stalls. The tracking guide covers how to run those check-ins without turning every meal into arithmetic.

How to use this

The macro calculator sets grams of protein, fat and carbohydrate for a calorie target, and the protein calculator sets the protein line from body weight. Those grams are what the hand measures above are standing in for.

If you want a number rather than a proportion, the calorie deficit calculator gives the target to aim at, and the tracking guide explains how loosely you can hit it before the deficit disappears.

Eating out, travel and buffets

Restaurant portions run one and a half to two times a home portion, and the gap is widest for starch and oil rather than for protein. A workable default is to take your own estimate and add half again for a restaurant, and to double it for anything deep fried or dressed at the table.

A 20 percent error on a 600 kcal meal is 120 kcal, which is small against the 500 kcal deficit being run and vanishes in a weekly average. What does not vanish is eating out five nights a week while estimating it as though it were home cooking.

Buffets defeat estimation altogether, so switch methods rather than trying harder. Fix the number of trips to the counter before starting, treat the first plate as the meal, and use a smaller plate. The exact calories are unknowable; the decision about the second plate is not, and that decision is where the calories are.

Frequently asked questions