Weight Loss Plateau: Four Causes and How to Check

A weight loss plateau is a stretch where the scale stops moving while the eating plan has not changed. The scale is a noisy instrument. Before changing anything, work out which of four things has happened: intake has drifted and is no longer measured, maintenance has fallen with body mass, non-exercise activity has fallen without notice, or water is masking fat loss. In most plateaus the first three are the cause, and the standard response — cut calories further — makes the problem worse.

The arithmetic below uses Mifflin-St Jeor for resting metabolic rate and the Atwater factors for energy in food. Both are named so the numbers can be repeated with different inputs.

The four causes, in order of how often they are the cause

1. Intake has drifted up and is no longer measured

This is the most common cause. A plan that was weighed for the first six weeks becomes estimated by week twelve. Cooking oil, the spoon of peanut butter, the handful of nuts, the second coffee with milk — each is small, and together they can add 200-400 kcal per day without any decision to eat more.

The diagnostic is a measured week. Weigh every item that has calories for seven consecutive days, including drinks, oils, and tastes while cooking, and log it. Mifflin-St Jeor gives the maintenance figure to compare against:

RMR (men)   = 10 x weight(kg) + 6.25 x height(cm) - 5 x age + 5
RMR (women) = 10 x weight(kg) + 6.25 x height(cm) - 5 x age - 161

Example: woman, 38, 165 cm, 78 kg
Step 1  RMR = (10 x 78) + (6.25 x 165) - (5 x 38) - 161
Step 2      = 780 + 1031.25 - 190 - 161
Step 3      = 1460 kcal/day
Step 4  Maintenance at light activity (x1.4) = 1460 x 1.4
Step 5      = 2044 kcal/day, band roughly 1840-2250

The 1.4 multiplier is a convention, not a measurement. Published error on Mifflin-St Jeor is around 10% for most adults, which is why the band is given rather than the midpoint. If the measured week averages within 100 kcal of the maintenance band, intake drift is not the cause and the next check applies.

2. Maintenance has fallen with body mass

Maintenance is not fixed. It falls as mass falls, and it falls faster than most people expect because both the RMR term and the activity term shrink. A 10 kg loss at constant height and age lowers RMR by about 100 kcal per day through the weight term alone, and the activity multiplier applies to a smaller base.

The diagnostic is to recompute maintenance at the current weight and compare it with the intake that produced the loss. If the deficit has closed, the scale stopping is arithmetic, not failure.

Same woman, now 68 kg (10 kg lost)
Step 1  RMR = (10 x 68) + (6.25 x 165) - (5 x 38) - 161
Step 2      = 680 + 1031.25 - 190 - 161
Step 3      = 1360 kcal/day
Step 4  Maintenance at x1.4 = 1360 x 1.4
Step 5      = 1904 kcal/day
Step 6  Previous maintenance was 2044 kcal/day
Step 7  Fall in maintenance = 2044 - 1904 = 140 kcal/day

An intake of 1,600 kcal produced a 444 kcal deficit at 78 kg and a 304 kcal deficit at 68 kg. The deficit narrowed by 140 kcal without any change in eating. The 7,700 kcal per kilogram figure often used to predict the timeline describes adipose tissue, not body weight, and it is wrong in two directions: early loss includes water and glycogen, so it overstates fat loss; late in a diet maintenance has fallen, so it overstates the rate. Both errors push the same way.

3. NEAT has fallen unconsciously

Non-exercise activity thermogenesis is the energy spent on everything that is not sleeping, eating, or deliberate exercise: walking, standing, fidgeting, taking stairs. It is the most variable component of daily expenditure and it falls during a deficit without any conscious decision. People sit sooner, take the lift, and move less in the evening.

The diagnostic is a step count and a standing time, compared with the same measure from the first weeks of the diet. A fall of 2,000 steps per day is roughly 80-120 kcal for most adults, which is enough to erase a small deficit. The check is cheap: the phone already counts steps.

4. Water retention is masking fat loss

Fat loss and water shifts are separate processes that the scale adds together. A high-salt meal, a new training block, hormonal cycle timing, or a change in carbohydrate intake can hold 1-2 kg of water for days to weeks. Fat can be leaving while the scale is flat.

The diagnostic is a waist measurement and a longer averaging window. A weekly average of daily weights, taken first thing after waking, removes most of the day-to-day noise. If the waist is falling while the weekly average is flat, fat is leaving and water is the explanation. This is the cause to check before changing anything, because acting on it means cutting calories for a problem that is already resolving.

What changes the figure

The maintenance figure is not one number. It depends on which equation produced it, which activity multiplier was applied, and which body mass was used. The table below gives the same woman at three points.

The doubly labelled water method measures expenditure directly rather than predicting it, and it is the reference standard. It is not available at home. The predicted figures carry roughly 10% error, and the activity multiplier carries more, because self-reported activity is unreliable in both directions.

What people get wrong about a plateau

The standard advice is to cut calories further. That is the wrong first move three times out of four, and the mistake is natural rather than stupid. The scale is the only feedback most people have, and a flat scale looks like a stalled deficit. Cutting is the action that follows from that reading.

But three of the four causes above are not fixed by cutting. If intake has drifted, cutting adds a restriction on top of an unmeasured surplus and the measured week is still the answer. If maintenance has fallen, cutting widens a deficit that may already be larger than intended. If NEAT has fallen, cutting does not restore movement and may reduce it further. Only water retention resolves on its own, and cutting in response to it means restricting for a problem that was already ending.

A second common error is treating a single weigh-in as data. Day-to-day weight varies by 1-2 kg from water alone. A plateau is a pattern over two to three weeks of weekly averages, not a reading from one morning.

A third is the assumption that the plan worked before, so it must still work. The plan did work. The body it was applied to has changed, and the maintenance figure that made it work has moved.

When to seek help

This page describes arithmetic, not medical care. A plateau is not a diagnosis and nothing here treats a clinical condition. Anyone restricting intake to the point of distress, losing control around food, or using the scale as a daily verdict should speak to a clinician. In the United States, the National Eating Disorders Association helpline is available by text or chat; in the United Kingdom, Beat runs a helpline for eating disorders. A plateau that persists for more than six weeks with a measured intake at or below the maintenance band, or that comes with fatigue, hair loss, or loss of menstrual cycles, is a reason to see a doctor rather than to cut further.

The order of checks

Run the measured week first. Recompute maintenance at the current weight second. Compare step counts third. Take a waist measurement and a two-week weekly average fourth. Only after all four have been checked is a change to intake the next question, and the answer is usually a smaller adjustment than the first instinct suggests.