How to set your calories without a coach

Every calorie calculator you have ever used gave you a confident number. None of them knew your metabolism. They knew your height, weight, age and a rough activity guess, and they ran that through an equation fitted to a population you are not necessarily representative of.

That does not make them useless. It makes them a starting point that needs correcting — and the correcting is the part nobody explains.

Step one: estimate maintenance

Maintenance is the number of calories that keeps your weight stable. Two ways to estimate it, both fine:

The quick version. Bodyweight in pounds × 14 to 16. Bodyweight in kilograms × 31 to 35. Use the lower end if you sit at a desk most of the day, the higher end if you are on your feet or physically active at work.

An 80 kg office worker: 80 × 31 = roughly 2,500 calories.

The slightly better version. The Mifflin-St Jeor equation estimates your resting metabolic rate, then you multiply by an activity factor. It is more precise on paper. In practice the activity multiplier introduces so much error that the quick version usually lands in the same place. Use whichever you prefer; do not agonise.

Either way, treat the result as a hypothesis.

Step two: subtract

For fat loss, take 20 to 25 per cent off maintenance.

Our 80 kg example at 2,500 maintenance lands around 1,900 to 2,000 calories.

That is a deficit of roughly 500 to 600 calories a day, which corresponds to somewhere around half a kilogram of fat loss a week. It is aggressive enough to see progress within a fortnight and moderate enough to be sustainable, to protect training quality, and to avoid the muscle loss that comes with harder cuts.

Why not cut harder

Because the thing you are optimising is not speed, it is the amount of fat you lose before you stop.

A 40 per cent deficit works brilliantly for three weeks. Then training performance falls off, hunger becomes genuinely difficult rather than merely inconvenient, sleep often worsens, and a larger share of what you lose comes from muscle rather than fat. Most people abandon it, regain, and conclude that dieting does not work for them.

There is also a specific trap in very low intakes: below roughly 1,200 calories for women or 1,500 for men, hitting protein and micronutrient targets becomes genuinely difficult. You are not just eating less energy, you are eating less of everything.

The deficit that produces the best twelve-month outcome is almost never the one that produces the best three-week outcome.

Step three: the correction that actually sets your calories

This is the part that matters, and it is the part that calculators cannot do.

Eat your target number for two full weeks. Weigh yourself every morning under the same conditions. Compare week two's seven-day average to week one's.

Then read the result:

• Losing 0.5 to 1 per cent of bodyweight per week — your number is right. Do not change anything.
• Losing nothing across two weeks of accurate tracking — your maintenance estimate was too high. Reduce by about 10 per cent and repeat.
• Losing more than 1 per cent a week — you are cutting too hard. Eat a bit more. Fast loss is not a win; it is a preview of a stall and a signal that lean tissue is going with it.

Two weeks, not one. A single week can be entirely masked by water retention, and acting on one week of data is how people end up eating 1,400 calories for no reason.

Why the scale needs an average

Daily bodyweight is noise. Salt, carbohydrate intake, water, hormones, training soreness and whatever is currently in your digestive tract can move you two kilos overnight while your actual weekly fat loss is 400 grams.

If you judge yourself on a single morning's number you will feel like you are failing about half the time, on a diet that is working perfectly. Weigh daily, but only ever compare weekly averages. The average is the signal; the individual readings are the noise it is buried in.

The part everyone gets wrong: tracking accuracy

Before you reduce your calories because "the deficit is not working", audit the tracking. Studies on self-reported intake consistently find people under-report by 20 to 50 per cent, and the people doing it are not lying — they genuinely do not see it.

The usual culprits:

• Cooking oil. A generous glug is 200 to 300 calories. Most people log it as nothing.
• Drinks. A latte, a couple of beers, a large orange juice. Several hundred calories that do not feel like eating.
• Bites, licks and tastes. Finishing a child's plate, testing the sauce, three crisps from the bag.
• Weekend drift. Tracking Monday to Friday accurately and eating freely on Saturday and Sunday can erase the entire week's deficit. Two days at +1,000 cancels five days at −400.
• Portion estimation. "A serving" of rice, pasta or nuts is usually considerably more than the packet's serving size.

Weighing your food for two weeks is unpleasant and extremely informative. You do not have to do it forever. You do it once, calibrate your eye, and then estimate with far better accuracy afterwards.

Do you have to count at all?

No. Counting is a measuring tool, not a requirement.

Plenty of people get lean on habit-based approaches: protein at every meal, vegetables at most, sensible portions, limited liquid calories, consistent meal timing. If that works for you, it works and there is nothing inferior about it.

The advantage of counting is diagnostic. When progress stalls, someone tracking can see exactly what changed; someone not tracking is guessing. A reasonable compromise is to track for a few weeks at the start, learn what your normal portions actually contain, and then stop.

One caution: if you have any history of disordered eating, calorie counting can be genuinely harmful, and habit-based approaches are not just an acceptable alternative but the better clinical choice. That is a conversation for a professional, not an article.

Recalculate as you go

Your maintenance falls as you lose weight. A smaller body burns fewer calories doing everything, including existing. The 2,500 that was maintenance at 80 kg might be 2,300 at 72 kg.

Recalculate roughly every 5 per cent of bodyweight lost, or every six to eight weeks. Otherwise your carefully constructed deficit slowly shrinks toward zero and you cannot work out why progress stopped.

What about macros beyond protein?

Once protein is set and total calories are set, the split between carbohydrate and fat matters far less than most people assume.

The evidence on this is reasonably clear: when calories and protein are matched, low-carbohydrate and low-fat diets produce very similar fat loss. The metabolic advantage claimed for one or the other has not held up in controlled feeding studies.

What differs is how each suits particular people. Some feel better and train harder on more carbohydrate. Others find higher fat more satiating and easier to sustain. Neither is superior; the one you can adhere to is the one that works.

Two floors worth respecting. Fat should not go below roughly 0.6 g per kilogram of bodyweight — it is required for hormone production and for absorbing fat-soluble vitamins. And carbohydrate should not go so low that your training collapses, since maintaining training quality is one of the four things protecting your muscle during a deficit.

Beyond those floors, split the remainder however you prefer. It is a preference question dressed up as a science question.

Deficits and diet duration

A deficit is a stressor, and stressors are best applied in blocks rather than indefinitely.

A sensible structure is eight to sixteen weeks in a deficit, followed by a period at maintenance of at least a few weeks before deciding whether to run another block. For someone with a lot to lose, this means several cycles rather than one enormous continuous diet — and the evidence on intermittent versus continuous restriction suggests the cycled approach produces comparable or better results, mostly because people actually finish it.

The signals that a block has run long enough: hunger that is getting worse rather than settling, training performance declining across weeks, sleep worsening, persistent cold, or simply the sense that adherence is becoming fragile. None of those improve by pushing harder.

The short version

Bodyweight in kilos × 31 to 35 for maintenance. Take 20 to 25 per cent off. Eat that for two weeks, weighing daily and comparing weekly averages. Adjust by 10 per cent if the average is not moving in the right direction, and audit your tracking before you assume the number is wrong.

Then leave it alone. The number is not the intervention — the consistency is.