How Many Calories Should I Eat to Lose Weight?

People usually ask "How many calories should I eat to lose weight?" as if there's one number out there with your name on it. There isn't, but there's a good starting point and a reliable way to correct it.

The short answer: eat about 300–500 calories a day below your maintenance, which for most people means losing 0.5–1% of their bodyweight a week. For a lot of adults that lands somewhere between 1,500 and 2,300 calories a day. Then weigh yourself for two to three weeks and let the trend tell you whether the estimate was right, because the first number you get from any calculator is a guess.

The rest of this post is how to make that guess a good one, and what to do when it's off.

Start with your maintenance calories

Maintenance (your TDEE, or total daily energy expenditure) is what you burn in a day: your resting metabolism, everyday movement like walking and standing, your workouts, and the energy it takes to digest food. Eat that much and your weight holds steady. Eat less and it drops.

You can't measure it at home, so you estimate it. Our TDEE calculator uses the Mifflin-St Jeor equation, or Katch-McArdle if you know your body fat percentage, then multiplies by an activity level. The good formulas land within about 10% of the real number for most people. That sounds precise until you do the math: 10% of 2,400 is 240 calories, which is half of a typical deficit.

Think of the estimate like the range your car shows on a full tank. It's a sensible number based on averages, but your real mileage depends on how you drive, and you only find out after a few tanks. Your first few weeks of eating and weighing are those tanks.

One opinion before you pick an activity level: if you're torn between two options, pick the lower one. People who sit at a desk and lift three times a week tend to see themselves as "moderately active", but an hour in the gym burns less than most people assume, and the other 23 hours matter more. Starting a little low costs you nothing. Starting high means weeks of wondering why nothing's moving.

How big should your deficit be?

Aim to lose 0.5–1% of your bodyweight a week. That scales sensibly with size: a bigger body can handle a bigger deficit without falling apart.

To turn that into calories, the usual rule is that a pound of fat holds about 3,500 calories (roughly 7,700 per kg). So losing 1 lb a week takes a deficit of about 500 calories a day, and 0.5 kg a week takes about 550.

A caveat on that rule: it's fine for planning the next month, but it overstates weight loss over longer stretches. As you get lighter you burn less, so the same intake produces a smaller deficit than it did at the start. That's normal, and it's why you'll re-check your numbers every few weeks rather than setting them once.

Why we'd go slower rather than faster

If you lift, we'd start at the slow end, somewhere around 0.5–0.75% a week. There's a Norwegian study of elite athletes (about two dozen of them) that compared losing roughly 0.7% versus 1.4% of bodyweight a week while they kept training. The slower group took longer to reach their target, but they came out of it with a little more lean mass and generally bigger strength gains, while the faster group's lean mass barely moved. It's a small study in athletes who train far more than most of us, so don't read too much into the exact numbers. But it matches what lifters commonly report: the faster the cut, the sooner your gym numbers start sliding.

The bigger reason is simpler. A deficit you can hold for 16 weeks beats one you abandon after four. Big deficits make you hungrier, make your workouts feel flat and make social meals a negotiation, and most people quietly drift off them.

Two exceptions. If you have a lot to lose, 1% a week (sometimes a bit more early on) is reasonable, because there's more stored energy to draw from. And if you're already fairly lean, go slower than 0.5%, since that's where muscle loss becomes a real risk.

Keep a floor in mind too: as a general guide, don't go below about 1,500 calories a day for men or 1,200 for women without medical supervision. If your math puts you under that, the fix is a slower rate, not a harder diet.

A worked example: a 180 lb lifter

Say you're a 35-year-old man, 5'10" (178 cm) and 180 lb (82 kg). You work at a desk and lift three times a week.

• Resting metabolism (Mifflin-St Jeor): about 1,760 calories.
• Maintenance at the "light" activity level: about 1,760 × 1.375 = 2,420 calories.
• Target rate: 1 lb a week, which is a little over 0.5% of 180 lb.
• Daily target: 2,420 − 500 = about 1,920 calories.

Then protein, at 0.85 g per pound: 180 × 0.85 = 153 g, or 612 calories. That leaves about 1,300 calories for carbs and fat. A balanced split works out to roughly 180 g of carbs and 65 g of fat. The macro calculator will do this for you, and it'll let you shift the split toward fewer carbs if that's how you prefer to eat.

So: 1,920 calories, 153 g protein, 180 g carbs, 65 g fat. That's the plan for the next three weeks, not forever.

Why protein comes first

When you eat less, your body is more willing to use muscle for energy. Plenty of protein plus lifting is the best protection you have, and protein keeps you fuller than the same calories of carbs or fat, which makes the deficit easier to live with.

Somewhere around 0.7–1 g per pound of bodyweight (1.6–2.2 g/kg) covers most people while dieting. If you have a lot of weight to lose, base it on your goal weight instead, or the number gets silly. The protein calculator gives you a figure and splits it across meals.

Once protein is set, how you divide the rest between carbs and fat matters much less than people argue about. Pick the split that fits how you like to eat. Low-carb works for some people because it cuts out the foods they overeat, not because of anything special about carbs.

Track honestly for the first few weeks

This is where most plans quietly fail. Not because people lie, but because nearly everyone underestimates what they eat.

The classic demonstration is a 1992 study in the New England Journal of Medicine. Researchers took a group of people with obesity who said they couldn't lose weight despite eating very little, then checked their food and exercise diaries against energy use measured with doubly labelled water (a lab method that tracks calories burned over days). On average they under-reported their food by 47% and over-reported their exercise by 51%. The group was small, and it was chosen precisely because its members felt "diet-resistant", so the typical gap for the rest of us is smaller. But a gap of a few hundred calories a day is common, and that's the whole deficit.

The usual leaks:

• Cooking oil. A tablespoon is about 120 calories, and a free pour is often two or three.
• Drinks. Lattes, juice, beer and wine rarely make it into the log.
• Bites while cooking and the last few forkfuls of someone else's plate.
• Sauces and dressings. Weigh them once and you'll never eyeball them the same way again.
• Weekends. More on this below.

Weigh your food at home for the first week or two. You don't need to do it forever. You're calibrating your eye, so that later "a portion of rice" means 150 g instead of whatever fits on the spoon.

It also helps to log as you eat rather than from memory at 10pm. In Helthy, one camera handles meal photos, barcodes and nutrition labels, and there's a quick-add for when you just know the calories. Use whichever is fastest for the meal in front of you, because the method you'll actually keep using is the accurate one.

The weekend problem

Here's how a reasonable plan disappears without anyone noticing. You hit your 500-calorie deficit Monday to Friday: that's 2,500 calories banked. Then Saturday is brunch and a few drinks, and Sunday is a takeaway, and each ends up about 1,000 calories over maintenance. That's 2,000 back. Your net for the week is 500 calories, or roughly 0.15 lb of fat.

You haven't done anything outrageous, and the scale hasn't moved. You don't have to give up weekends. But log them with the same honesty as Tuesday, and if they're consistently big, either plan for them (eat a bit less on weekdays) or accept a slower rate.

Don't eat back your watch calories

This one's an opinion, and some coaches disagree. If your maintenance estimate already includes your training (which it does if you picked an activity level), adding the calories your watch says you burned counts that exercise twice. Burn estimates from wrist trackers are also some of the least reliable numbers they produce. We'd leave them out of your food budget entirely and let your weight trend decide whether you need more.

How to tell if it's working

Weigh yourself every morning, after the bathroom and before eating, and look only at the weekly average.

Daily weight swings by 1–2 kg (2–4 lb) from water, salt, carbs, hormones and what's still in your gut. A single weekly weigh-in can land on a bad day and send you in the wrong direction. Seven readings averaged together smooth that out. We think daily weighing is the better habit for most people, provided you can treat the single number as noise. If one reading can ruin your morning, weigh less often and accept a slower answer.

Expect the first week to look better than it is. Cutting calories (especially carbs) drains some stored glycogen, which holds water, so a drop of 2–4 lb in week one is common and mostly water. Judge the rate from week two onwards.

Adjusting from real data

Back to the 180 lb lifter, aiming for 1 lb a week on 1,920 calories. His weekly averages:

He's losing 0.4 lb a week instead of 1 lb. The gap is 0.6 lb a week, which is about 0.6 × 3,500 ÷ 7 = 300 calories a day. Either he's eating about 300 more than he logs, or his real maintenance is about 300 lower than the formula said. Probably a bit of both.

The fix, in order:

1. Tighten the logging for a week: oil, drinks, weekends, the bites while cooking.
2. If that changes nothing, drop 150–200 calories a day and give it another two to three weeks.
3. If you're losing much faster than planned (and your training is suffering), add 150–200 back.

Small changes, then wait. The most common mistake after "not tracking accurately" is changing the plan every four days based on noise.

You'll need to do this more than once. As your weight falls, your maintenance falls with it, so a target that worked in month one will slow down by month three. Helthy re-checks your maintenance every week from your steps and logged workouts, and with Helthy Pro, Smart Calories also adjusts it from your weight trend and what you've eaten, which is the same comparison as the table above, done for you.

When the number isn't the problem

If you've tracked carefully for a month, cut twice, and your weekly average still hasn't moved, it's worth a look beyond the diet. Some medical conditions and medications affect weight and appetite. If that sounds like you, a GP visit is the sensible next step, rather than dropping calories again.

For most people, though, it really is the arithmetic, done honestly and given enough time. If you're a lifter with 10–30 lb (5–14 kg) to lose, start at about 0.5–0.75% of your bodyweight a week, keep protein near 0.85 g per pound, and don't touch the plan for three weeks. Go faster only if you have a lot to lose, and slower once you're lean enough that your gym numbers start slipping.