Calories13 min read

How Big Should Your Calorie Deficit Be? Why the 3,500-Calorie Rule Overshoots, and What to Expect Instead

The 3,500-calorie rule overshoots: Hall’s model says a 10 kcal/day cut means about 1 lb of eventual loss, half within a year. Here’s how big your deficit should be, safe minimums and what to expect.

How Big Should Your Calorie Deficit Be? A notebook with a weekly weight trend line beside a plate of balanced food and a glass of water — MacrosMeasure
How Big Should Your Calorie Deficit Be? A notebook with a weekly weight trend line beside a plate of balanced food and a glass of water — MacrosMeasure
D
Dr. James Whitfield, RD
Registered Dietitian & Sports Nutritionist, MS

In this guide

  1. The rule everyone knows (and trusts)

  2. Where 3,500 comes from

  3. Why it overshoots: you change while you diet

  4. The better rule: 10 calories per pound

  5. What to expect month by month

  6. How big should your deficit be?

  7. The pace that protects muscle: 0.5–1% a week

  8. The minimum-calorie question

  9. Fast versus slow: does pace affect regain?

  10. When to recalculate

  11. Stalls versus real plateaus

  12. A note for South and East Asian dieters

  13. A four-week plan

  14. The bottom line

  15. Frequently asked questions

Here's the weight-loss rule almost everyone has heard: cut 500 calories a day, lose a pound a week. Cut 1,000, lose two. It's printed on health-agency websites, quoted by personal trainers and baked into countless calorie apps. It sounds so tidy that it feels like physics.

It isn't physics. It's arithmetic that pretends your body stays the same while you diet. In 2011, a team led by Kevin Hall at the US National Institutes of Health published a model in The Lancet showing that the rule overshoots badly over months and years, and that official bodies, including the UK's NHS and the US NIH, had been repeating it. The model behind that paper powers the NIH's Body Weight Planner and answers the question most dieters actually have: if I cut this much, what will really happen, and when? We read the paper, the guidance from health agencies and the trials on pace and regain, and built a practical guide to choosing a deficit that works.

The short version: the 3,500-calorie rule treats weight loss as a straight line. The dynamic model says that for an average adult with overweight, every sustained 10 kcal/day change in intake leads to about 1 lb (0.45 kg) of eventual weight change, with half of it arriving in about a year and 95% in about three years. So a 500-calorie cut isn't "a pound a week forever". It's closer to a pound a week at the start, slowing as your body gets smaller and burns less. For most people, a deficit of 10–25% below maintenance, losing 0.5–1% of body weight per week, balances progress with muscle, energy and sanity. Very low intakes (under about 800 kcal) belong under medical supervision.

The Rule Everyone Knows (and Trusts)

The idea is this: a pound of body fat holds about 3,500 calories (or a kilogram holds about 7,700), so if you eat 500 fewer calories than you burn each day, you'll lose 3,500 calories a week, which is a pound. Repeat for a year and you lose 52 pounds. At that pace a 100 kg person would vanish in about two years.

Everyone who has dieted knows that doesn't happen. Weight comes off briskly in the first few weeks, then slows to a crawl even though you haven't changed what you do. That isn't a failure of willpower. It's the rule failing, not you.

Where 3,500 Comes From

The number comes from a 1958 estimate by Max Wishnofsky of the energy stored in a pound of body fat, which is mostly fat with a little water and protein. As a statement about the energy content of a pound of fat, it's roughly right. As a statement about how weight changes over time, it's badly wrong, because it makes two silent assumptions:

  1. The weight you lose is all fat. In reality, some lean tissue and water go too, and they carry fewer calories per kilogram.
  2. Your energy needs stay fixed. The rule assumes you burn the same calories at 80 kg as you did at 100 kg.

The second is the killer.

Why It Overshoots: You Change While You Diet

When you lose weight, your body gets cheaper to run. A smaller body has less tissue to maintain, burns fewer calories moving and needs less energy for digestion. Your maintenance calories fall as you lose, so the gap between what you eat and what you burn shrinks, even if your food stays exactly the same.

Imagine an adult who burns 2,500 calories a day and cuts to 2,000. The gap is 500. After losing 10 kg they might burn 2,300 a day, so the same 2,000 is now a 300-calorie gap. After 20 kg, maybe 2,100, and the gap is 100. Weight loss slows because the deficit is being gradually eaten up by your own smaller size, not because your metabolism has "broken". In Hall's dynamic model, the deficit eventually closes entirely and weight settles at a new, lower plateau.

The Better Rule: 10 Calories per Pound

From the model, Hall and colleagues proposed a rule of thumb for an average adult with overweight: every sustained change of 100 kJ per day in energy intake leads to an eventual body-weight change of about 1 kg, equivalently 10 kcal per day per pound. About half of that change arrives in roughly a year, and about 95% in roughly three years.

Sustained cut in intakeEventual weight loss (rule of thumb)Roughly halfway after
100 kcal/dayAbout 10 lb (4.5 kg)About 1 year
250 kcal/dayAbout 25 lb (11 kg)About 1 year
500 kcal/dayAbout 50 lb (23 kg)About 1 year
750 kcal/dayAbout 75 lb (34 kg)About 1 year

This is an approximation for an average adult with overweight and moderate, sustained changes. Individuals vary, and leaner people lose less per calorie. It also assumes you actually keep eating that much less.

The paper's own example makes the contrast vivid. For a 100 kg adult cutting roughly 480 calories a day, the old rule predicts around 22 kg lost in the first year and keeps going. The dynamic model predicts around half that in year one, flattening out at roughly 75 kg after about three years. Neither is wrong about the first week. The old rule is badly wrong about month six.

What to Expect Month by Month

TimeframeWhat usually happens on a steady, moderate deficit
Weeks 1–2Fast drop on the scale, partly water and glycogen as you eat less food and carbohydrate
Weeks 3–12Steadier fat loss; the average pace is close to what the simple rule predicts at first
Months 4–6Pace slows as your body shrinks and your maintenance falls; this is normal
Months 6–12Approaching about half of the eventual change; time to recalculate your targets
Year 1–3Weight approaches a new plateau; the deficit has largely closed

This is why a plan that promises a pound a week forever is selling a fantasy, and why the useful question isn't "how fast can I go?" but "how big a deficit can I hold, comfortably, for months?"

How Big Should Your Deficit Be?

Start with your maintenance calories from the TDEE calculator, then choose a deficit as a percentage. Most dietitians use 10–25%, with the middle of the range the best balance for most people.

Maintenance (TDEE)10% deficit15% deficit20% deficit25% deficit
1,600 kcal1,4401,3601,2801,200
2,000 kcal1,8001,7001,6001,500
2,500 kcal2,2502,1252,0001,875
3,000 kcal2,7002,5502,4002,250
DeficitBest forTrade-offs
10–15%Lean people, people who train hard, those wanting a sustainable pace, anyone with a low TDEESlower; needs patience
15–20%Most people with moderate weight to loseSome hunger; protein and sleep matter
20–25%People with more to lose who can tolerate it, in time-limited blocksHigher risk of muscle loss, fatigue and rebound if held too long
Above 25%Usually supervised, short-term plansHarder to meet protein and nutrient needs; not for most people

You can let the calorie deficit calculator do the sums, and the calorie calculator will give you a full target.

The Pace That Protects Muscle: 0.5–1% a Week

A more useful measure than "pounds a week" is percent of body weight per week, because it scales with your size. A common guide is 0.5–1%. Smaller or leaner people sit at the lower end, larger people can tolerate the upper end.

Body weight0.5% per week1% per week
60 kg (132 lb)0.3 kg (0.7 lb)0.6 kg (1.3 lb)
70 kg (154 lb)0.35 kg (0.8 lb)0.7 kg (1.5 lb)
80 kg (176 lb)0.4 kg (0.9 lb)0.8 kg (1.8 lb)
90 kg (198 lb)0.45 kg (1.0 lb)0.9 kg (2.0 lb)
100 kg (220 lb)0.5 kg (1.1 lb)1.0 kg (2.2 lb)

These are targets for the average pace over several weeks, and the first fortnight will often run faster. Use the weekly average of your daily weigh-ins, not a single morning. To protect muscle while you do it, keep protein up (roughly 1.2–1.6 g per kg of body weight, using the protein intake calculator) and keep lifting. A meta-analysis by Hudson and colleagues (Advances in Nutrition, 2020) found that protein around 1.3 g/kg reduced lean-mass loss during energy restriction.

The Minimum-Calorie Question

"What's the lowest I can safely eat?" has no single answer, but there are widely used guardrails:

GuidanceTypical figure
Common self-directed minimums in US public-health materials (NIH/NHLBI)About 1,200–1,500 kcal a day for women and 1,500–1,800 for men, adjusted for size and activity
Low-energy and very-low-energy diets (800 kcal and under)Medically supervised only; usually time-limited and with nutritionally complete products
A personal checkStaying near or above your estimated BMR for long periods is a sensible guardrail for most people, though BMR estimates themselves are uncertain

Two cautions. First, these floors don't apply to children, teenagers, pregnant or breastfeeding women, older adults or people with medical conditions. Second, eating very little makes it hard to get enough protein, fibre and micronutrients, and makes binge-and-rebound cycles more likely. If your TDEE is low, say 1,600, a 25% deficit takes you to 1,200 straight away, which is a reason to pick a smaller percentage, not a reason to drop further.

Fast Versus Slow: Does Pace Affect Regain?

It's widely believed that fast weight loss "always comes back". The evidence is more nuanced. In a randomised trial by Purcell and colleagues (The Lancet Diabetes & Endocrinology, 2014) of about 200 adults with obesity, people who lost weight rapidly over 12 weeks and those who lost it gradually over 36 weeks regained similar amounts over the following three years. A similar 2016 trial reached a comparable conclusion. And the DiRECT trial in the UK showed that a supervised low-calorie programme could put type 2 diabetes into remission for many people.

What that means in practice:

  • Pace itself may matter less than what you do afterwards. Maintenance habits, protein, strength training, regular weighing and ongoing support predict regain better.
  • Fast loss has real costs. More muscle loss, more hunger, nutrient gaps and difficulty adhering, particularly outside supervised programmes.
  • For most people, a moderate pace is the best fit, not because it's magic, but because it's easier to sustain and easier to do safely.

When to Recalculate

Because your maintenance falls as you lose, your targets should move too. A simple rule: recalculate every time you lose about 5% of your body weight, or every 6–8 weeks. Using the 10 kcal-per-pound rule, each kilogram you lose lowers your maintenance by roughly 22 calories a day, so after 10 kg your original 500-calorie deficit has shrunk to around 280 if you haven't adjusted.

When you recalculate, keep the deficit as a percentage of your new TDEE rather than chasing a fixed number, and don't be tempted to keep cutting food just because the deficit has narrowed. If you need to speed up, add movement first, or take a diet break.

Stalls Versus Real Plateaus

  • Normal fluctuation. Day-to-day weight moves by 1–2 kg with water, salt, carbohydrate and bowel contents. Judge by weekly averages.
  • A stall of 1–2 weeks is common, especially if you've been training harder or are retaining water. Check your waist and strength before changing anything.
  • A true plateau means no change in your weekly average for three to four weeks. The usual causes are drifting portions, "weekend" calories, hidden calories in oils and drinks, or an out-of-date TDEE. Our guide to eating the wrong number of calories covers the audit.
  • The fix is usually accuracy and a small adjustment (100–200 kcal or more steps), not a dramatic cut.

A Note for South and East Asian Dieters

Two things from this series compound here. First, as we explain in our BMR article, the common equations tend to overestimate resting metabolism in Indian and Chinese adults, by roughly 7–11%, so your calculated TDEE may be too high, and your "500-calorie deficit" may be much smaller in practice. Second, health risk rises at lower BMI and waist sizes in these groups, so modest weight loss from a lower starting point can still matter.

The practical advice: start from a TDEE 5–10% lower than the calculator gives (a rule of thumb, not a validated correction), hold for three weeks, and let your weekly average weight tell you whether to adjust. Don't respond to a stall by dropping calories far below the guardrails above. Raise protein, add walking and lifting, and recalibrate.

A Four-Week Plan

  1. Week 0: estimate maintenance with the TDEE calculator, adjust for the Asian-overshoot note if relevant, and choose a 15–20% deficit.
  2. Weeks 1–2: hit your calories and protein target on most days, and weigh yourself daily. Don't judge by the first week.
  3. Week 3: compare the weekly averages. Aim for 0.5–1% of body weight a week on average.
  4. Week 4: if the pace is under 0.3% for two weeks, audit intake first and recheck your TDEE. If it's over 1%, add 100–200 calories to protect muscle and energy.
  5. Every 6–8 weeks: recalculate maintenance and consider a one-week "maintenance break" at the new TDEE if you're tired, hungry or stalling.

The Bottom Line

The 3,500-calorie rule isn't wrong about the energy in a pound of fat. It's wrong about time. Your body shrinks as you diet, burns less, and closes the gap you opened, so the plain "pound a week" line bends into a curve that flattens over about three years. A better rule of thumb is 10 calories per day per pound of eventual change, with half arriving in about a year.

So choose a deficit you can hold: 10–25% below maintenance, around 0.5–1% of body weight a week, with protein high, strength training in and floors respected. Recalculate as you lose, expect the slowdown and judge by the trend. The aim isn't the biggest cut. It's the one you can still be doing in month six.

Frequently Asked Questions

Is the 3,500-calorie rule wrong?

It's roughly right for the energy in a pound of fat but wrong as a prediction of weight change over time, because it ignores that your energy needs fall as you lose weight. A 2011 Lancet study by Kevin Hall and colleagues showed it overestimates long-term loss.

How much weight will I lose with a 500-calorie deficit?

About a pound a week at the start, slowing over time. By the dynamic rule of thumb, a sustained 500 kcal/day cut leads to about 50 lb of eventual loss in an average adult with overweight, with about half in the first year and 95% by about three years.

What is Hall's rule of thumb?

Every sustained 10 kcal/day change in intake leads to about 1 lb (0.45 kg) of eventual weight change in an average overweight adult. Half of the change arrives in about a year and 95% in about three years.

How big should my calorie deficit be?

For most people, 10–25% below maintenance, with 15–20% a good middle for those with moderate weight to lose. A common pace target is 0.5–1% of body weight per week.

What is the minimum calories I should eat?

Common self-directed guidelines suggest about 1,200–1,500 kcal for women and 1,500–1,800 for men, adjusted for size and activity. Very low-calorie diets of 800 kcal or less should be supervised by a clinician.

Do fast diets cause more regain?

Not necessarily. A randomised trial found people who lost weight rapidly and gradually regained similar amounts over three years. Fast diets have other costs, such as muscle loss and nutrient gaps, and aren't suitable for everyone.

Why did my weight loss slow down?

Because your body is smaller and burns fewer calories, the gap between intake and expenditure shrinks. Recalculate your TDEE every time you lose about 5% of your body weight.

Should I cut calories further when I stall?

Usually not. Check accuracy first (portion drift, hidden calories, an out-of-date TDEE), add movement or protein, and consider a short maintenance break before cutting further.


References & Sources

  1. Hall, K. D. et al. (2011). Quantification of the effect of energy imbalance on bodyweight. The Lancet, 378(9793), 826–837.
  2. Hall, K. D. (2008). What is the required energy deficit per unit weight loss? International Journal of Obesity, 32(3), 573–576.
  3. Wishnofsky, M. (1958). Caloric equivalents of gained or lost weight. American Journal of Clinical Nutrition, 6(5), 542–546.
  4. Purcell, K. et al. (2014). The effect of rate of weight loss on long-term weight management: a randomised controlled trial. The Lancet Diabetes & Endocrinology, 2(12), 954–962.
  5. Hudson, J. L. et al. (2020). Protein intake greater than the RDA differentially influences whole-body lean mass responses to purposeful catabolic and anabolic stressors: a systematic review and meta-analysis. Advances in Nutrition, 11(3), 548–558.
  6. Lean, M. E. J. et al. (2018). Primary care-led weight management for remission of type 2 diabetes (DiRECT). The Lancet, 391(10120), 541–551.
  7. National Heart, Lung, and Blood Institute. Aim for a healthy weight: calorie guidance for weight loss.
  8. National Institute for Health and Care Excellence (2025). Overweight and obesity management (NG246).
  9. MacrosMeasure (2026). Free Calorie Deficit Calculator.
  10. MacrosMeasure (2026). Free TDEE Calculator.
Put it into practice

Calculate your own numbers — free.

18 science-backed calculators covering everything in this article and more.

🔥 Calorie Calculator 📊 TDEE Calculator All 18 calculators
How Big Should a Calorie Deficit Be? The 3,500 Rule Is Wrong