How to Break a Weight Loss Plateau

TL;DR: A plateau has three possible causes and you should test them in order. First, the window is too short: ten flat days on a noisy scale is not a plateau. Second, tracking drift, which is by far the most common real cause. Lichtman et al. (1992) measured a 47 percent gap between what "diet-resistant" subjects reported eating and what doubly labeled water said they actually ate. Third, genuine metabolic adaptation, which is real and measurable but smaller than the internet claims. Fix them in that order and most plateaus resolve without cutting calories at all.

---

You've been in a deficit for weeks. The scale hasn't moved in ten days. Same food, same training, same routine. Nothing's working.

Before you decide your metabolism is broken, understand that a stalled scale has three possible explanations, and they aren't equally likely. Two of them are boring. Only the third one is physiology. Work through them in order, because the fix for one will do nothing for the others.

First: is it actually a plateau?

Ten days is not enough data to call anything.

Bodyweight swings 2 to 4 lbs day to day on glycogen, sodium, fiber load, and hydration alone. If you're losing 0.75 lbs a week, that signal sits buried under noise several times its size. Two weeks of daily weigh-ins is roughly the minimum before a trend means anything, and you need a 7-day rolling average to see it. One Monday morning number compared against last Monday tells you almost nothing.

Three things reliably fake a plateau:

• A new training block. More volume means more muscle glycogen and the water bound to it. That can mask two weeks of real fat loss.
• Creatine, a salty weekend, or a bad night of sleep. All shift water. None shift fat.
• The luteal phase. Water retention here can hide fat loss for a full week or more, which is why women should compare cycle to cycle rather than week to week.

This is the same reason the scale lies about water weight and why daily weigh-ins beat weekly ones. More data points, averaged. Not fewer.

Before you change anything, check the other instruments. If the tape is moving, your lifts are climbing, and your progress photos look different, you aren't plateaued. You're recomposing, and the scale is the wrong tool for that job.

Second: tracking drift

If the trend really is flat over three weeks, the next suspect is your logging. Not your metabolism.

Lichtman and colleagues (1992) took ten subjects who believed they were diet-resistant, confirmed every one of them had a normal metabolic workup, then measured actual intake with doubly labeled water over 14 days. Reported intake averaged 1,028 calories. Measured intake was 2,081. That's a 47 percent underreport, and the same subjects overestimated their exercise by 51 percent. The error compounded on both sides of the equation.

These people weren't lying. Dhurandhar et al. (2015) made the broader point in a review of energy balance measurement: self-reported intake isn't a slightly noisy estimate of real intake, it's a systematically biased one, and the bias grows as intake grows.

Portions creep. Oils go unlogged. Bites while cooking never get counted. The usual culprits:

• Cooking oil. A "light drizzle" is 100 to 150 calories, and most people log zero.
• Nut butter. Heaping versus level tablespoons is a 50 to 100 calorie swing per serving.
• Restaurant meals. Routinely 30 to 60 percent above what you'd guess.
• Alcohol. Forgotten more often than any other category.
• Raw versus cooked entries. Picking the wrong one is a 25 percent error on rice and chicken. If you aren't sure which you should be using, here's the rule.

The test is one honest week. Weigh everything on a food scale, log the oil, log the bites, no estimating. Most people find their real deficit was 100 to 300 calories smaller than the number in their app. If weight starts moving, you found it, and you never needed to eat less. You needed to measure better. This is the single biggest source of error in calorie tracking.

Third: genuine metabolic adaptation

If your logging is airtight and the trend is still flat after three weeks, now you're looking at adaptive thermogenesis. It's real, it's well documented, and it's smaller than the "starvation mode" version people imagine.

The measured effect: total energy expenditure falls roughly 15 percent after a 10 percent weight loss, and part of that drop is larger than the loss of tissue alone would predict. Mechanically it shows up as reduced NEAT (you fidget and walk less without noticing), lower T3, suppressed leptin, and a smaller thermic effect of food because you're simply eating less food.

Two of those deserve emphasis. NEAT is the biggest and most underrated: unconscious daily movement can swing several hundred calories, and it falls quietly during a long deficit. That's why step count is worth defending when the scale stalls. Leptin suppression is the other one, and it's why hunger climbs faster than your body actually shrinks. The full mechanism is in metabolic adaptation during dieting.

Your body isn't broken. It's doing exactly what it evolved to do.

Three fixes for a genuine plateau

Recalculate TDEE at your current weight. This goes first because it's often the whole answer. Your maintenance at 200 lbs isn't your maintenance at 182 lbs. A deficit you set four months ago has quietly become maintenance. Rerun the numbers against today's bodyweight and set the deficit from there. Here's how to calculate maintenance calories properly.

Cut another 100 to 200 calories. Small and incremental. Don't drop 500 at once, and don't go below roughly 1,200 for women or 1,500 for men, where muscle loss and further suppression both accelerate.

Run a structured diet break. The MATADOR trial (Byrne et al., 2018) put 51 men with obesity through 16 weeks of energy restriction, delivered either continuously or as 2-week blocks alternating with 2-week blocks at energy balance. The intermittent group lost more fat, kept more fat-free mass, blunted the drop in resting energy expenditure, and held their loss better at six months. A break means two weeks at true maintenance, not a free-for-all. You'll gain 1 to 3 lbs of water and glycogen, and it comes back off in days. Here's how to structure one.

If you're on a GLP-1

Plateaus on semaglutide or tirzepatide follow a different curve, and treating one like a tracking problem is a mistake. Loss on these drugs typically flattens somewhere around the 12 to 18 month mark, as dose escalation ends and expenditure adapts downward. That adaptation gets compounded by lean mass loss, which can run a quarter to a third of total weight lost when protein and resistance training are neglected.

The intervention there isn't a bigger deficit. It's protecting the tissue that sets your metabolic rate: high protein, hard resistance training, and patience. We covered muscle loss on GLP-1s separately, because it's the failure mode that turns a plateau into regain.

What not to do

Slash calories hard. Going 1,800 to 1,200 overnight deepens adaptation and costs you muscle, which lowers your maintenance further. You'll buy two weeks of scale movement and pay for it for months.

Bolt on cardio. NEAT compensates. Add 45 minutes of steady state and your body quietly removes an equivalent amount of unconscious movement from the rest of your day. Cardio has a place, but it's a poor tool for prying open a stalled deficit.

Wait it out. A genuine plateau that has survived three to four weeks of accurate tracking will not resolve on its own. Something has to change.

Bottom line

Check your measurement window, then audit your logging, then intervene. Most plateaus die at step two, and the ones that survive to step three usually need a recalculated TDEE rather than heroic self-denial. Adaptation is real. It's also a couple hundred calories, not a broken metabolism.

Run your current stats through the free Cut Calculator to reset your deficit, or see where an honest rate of loss actually lands you with the Weight Loss Timeline. If you'd rather not redo this math every six weeks, Protokl recalculates your TDEE from your own logged intake and weight trend, so the deficit moves when your metabolism does.