Somewhere between a third and two-thirds of the weight you lose in the first week or two of a diet is water, not fat. The exact proportion depends heavily on what kind of diet you follow and how drastically you cut calories. A classic study of obese subjects found that on an 800-calorie ketogenic diet, about 61% of weight lost was water, while on an 800-calorie mixed diet with normal carbohydrate levels, water accounted for roughly 37% of the loss.1JCI Insight. Composition of weight lost during short-term weight reduction. Metabolic responses of obese subjects to starvation and low-calorie ketogenic and nonketogenic diets That range tells you something important: the dramatic number on the scale during your first week is real weight leaving your body, but most of it will come back the moment you return to normal eating.
Why Your Body Dumps Water First
The main driver is glycogen, a form of stored carbohydrate packed into your liver and muscles. Your body keeps a few hundred grams of glycogen on hand as quick-access fuel. The crucial detail is that glycogen doesn’t sit there dry. Every gram of glycogen binds to roughly three to four grams of water.2PubMed. Glycogen storage: illusions of easy weight loss, excessive weight regain, and distortions in estimates of body composition So if you have 400 grams of glycogen stored, that’s carrying along 1,200 to 1,600 grams of water, bringing the total package to somewhere around 1.6 to 2 kilograms (roughly 3.5 to 4.4 pounds).
When you start eating fewer calories than you burn, your body taps into glycogen before it gets serious about burning fat. As the glycogen drains, the water bound to it goes with it. Exercise research confirms this ratio: for every gram of glycogen replenished in muscle, at least three grams of water came along for the ride, and under full hydration conditions the ratio climbed even higher.3PubMed. Relationship between muscle water and glycogen recovery after prolonged exercise in the heat in humans The process works in reverse too, which is why refeeding after a diet causes that disheartening spike on the scale.
Liver glycogen tends to deplete relatively quickly. During fasting, the liver’s stores can run low within about 12 hours after your last meal, which is one reason an overnight fast already nudges the scale downward by morning.4PubMed Central. Flipping the Metabolic Switch: Understanding and Applying the Health Benefits of Fasting Muscle glycogen takes longer to deplete because your muscles are less willing to give up their reserves unless you’re actually using them through exercise.
How Diet Type Changes the Ratio
Not all diets produce the same split between water loss and fat loss. The study comparing ketogenic versus mixed diets at the same calorie level found a stark difference. On the ketogenic version (very low carbohydrate), subjects lost an average of about 467 grams per day, but only 35% of that was fat. On the mixed diet with conventional carbohydrate and fat proportions, they lost less total weight per day (about 278 grams), but nearly 60% of it was actual fat.1JCI Insight. Composition of weight lost during short-term weight reduction. Metabolic responses of obese subjects to starvation and low-calorie ketogenic and nonketogenic diets Protein loss was small in both cases, around 3-4%.
The reason is straightforward. When you eat almost no carbohydrates, your body burns through its glycogen stores faster and more completely. That forces out more water sooner, making the early scale drop look more impressive. When you eat a mixed diet with moderate carbohydrates, your body keeps partially restocking glycogen from the carbs you’re still eating, so less water gets flushed and a higher percentage of each lost gram is actual fat.
This is the core of a common frustration with low-carb diets: people see five or six pounds drop in the first week and feel elated, then watch the rate slow to a crawl in weeks two and three. They haven’t “stalled.” They’ve simply used up the easy water losses, and the remaining weight loss is now coming predominantly from fat, which is a much denser energy source and therefore burns slower. On the flip side, someone on a moderate calorie deficit with normal carb intake may see a more modest two to three pounds in the first week but find the rate stays more consistent over time.
The Insulin and Sodium Connection
Glycogen isn’t the only reason for early water loss. Your kidneys also start shedding sodium and water when insulin levels drop, and insulin drops whenever you eat less, particularly when you eat fewer carbohydrates. Animal research has shown that the normal rise in insulin after a meal helps the kidneys hold onto sodium, and when that insulin signal weakens, the kidneys let sodium and the water that follows it flow out more freely.5PubMed Central. The normal increase in insulin after a meal may be required to prevent postprandial renal sodium and volume losses
This is why people on very low-carb or fasting regimens sometimes feel lightheaded in the first few days. It isn’t just low blood sugar; it’s a genuine drop in blood volume from sodium and water excretion. The effect is temporary as the body adjusts, but it contributes meaningfully to that first-week scale number.
Salt intake itself also matters independently. Reducing dietary sodium causes the body to shed retained fluid, sometimes quickly. Hospitalized patients with chronic kidney disease who were placed on salt-restricted diets showed rapid weight loss within days, and the investigators attributed this largely to reduced fluid retention rather than any loss of muscle or fat.6PubMed Central. Rapid weight loss with dietary salt restriction in hospitalized patients with chronic kidney disease For people without kidney disease, the effect is smaller but still present. Data from the DASH-Sodium trial found that reducing sodium intake decreased both thirst and urine volume, consistent with the body finding a new, lower equilibrium for fluid balance.7PubMed Central. Effects of Sodium Reduction on Energy, Metabolism, Weight, Thirst, and Urine Volume: Results From the DASH-Sodium Trial So if your new diet also happens to cut out processed foods (which are loaded with sodium), you’re losing water from two directions at once: glycogen depletion and sodium reduction.
Why the First Week Feels So Dramatic
There’s a well-known rule of thumb that says you need a 3,500-calorie deficit to lose one pound. That rule was never great, but it’s especially misleading during the first few weeks. Because so much of early weight loss is water, which has essentially zero caloric cost to shed, the energy content of each pound lost early on is much lower than 3,500 calories. One study measured the actual energy content of weight change at week four and found it was about 2,200 calories per pound, far below the traditional estimate.8PubMed Central. Time to Correctly Predict the Amount of Weight Loss with Dieting
This matters because it explains a mismatch many people experience. In week one, a 500-calorie-per-day deficit might produce three or four pounds of loss. By month two, that same deficit produces something closer to one pound per week, maybe less. The deficit didn’t shrink (though metabolic adaptation does play a small role). What changed is that the body ran out of glycogen-bound water to dump, and each subsequent pound of loss is now mostly fat and therefore “costs” more in calories. Researchers have described this early phase as having a half-life of less than one week, meaning it’s intense but short-lived, typically wrapping up within four to six weeks.9PubMed Central. Energy content of weight loss: kinetic features during voluntary caloric restriction
This is worth understanding because it reframes what “real” progress looks like. The person who loses eight pounds in week one and one pound in week three hasn’t failed. They’ve transitioned from water-dominant loss to fat-dominant loss, which is exactly what a sustainable diet is supposed to do.
What Prolonged Fasting Reveals About the Breakdown
Longer fasting studies offer a sharper look at where weight loss comes from over time. A prospective trial of healthy men undergoing extended voluntary fasting tracked the composition of their lean soft tissue losses in detail. The results split into three categories: about 44% of the lean tissue lost was extracellular water (the fluid between cells and in the bloodstream), roughly 14% was glycogen and its associated water, and the remaining 42% was metabolically active tissue, meaning actual cellular material including some muscle protein.10PubMed Central. Is muscle and protein loss relevant in long-term fasting in healthy men? A prospective trial on physiological adaptations
A few things stand out here. First, water in its various forms (extracellular fluid plus glycogen-bound water) accounted for well over half of all lean tissue lost. Second, the glycogen-water component, at 14%, was relatively small by this point in a long fast because the glycogen was used up early. The extracellular water loss, at 44%, was the bigger player, reflecting the sustained drop in fluid retention from hormonal and sodium changes that persist beyond the glycogen phase. Third, actual muscle protein loss was real but not as catastrophic as people sometimes fear, at least in the context of a medically supervised setting.
Hormonal Fluctuations That Confuse the Scale
For anyone tracking weight closely, daily fluctuations of one to four pounds are normal and overwhelmingly driven by water, not fat. One source of confusion that affects roughly half the population is the menstrual cycle. A year-long prospective study of ovulating women found that fluid retention scores peaked on the first day of menstrual flow and were lowest during the mid-follicular phase (roughly a week after a period starts). Fluid retention then gradually climbed again around ovulation.11PubMed Central. Fluid Retention over the Menstrual Cycle: 1-Year Data from the Prospective Ovulation Cohort
This cyclical pattern means that a woman who starts a diet during the luteal phase (the two weeks before a period) may see very little change on the scale for the first week or two, even if she’s losing fat at a perfectly normal rate. The fat loss is being masked by rising fluid retention. Then, as her period arrives and the fluid shifts back, she might suddenly see a three-pound drop overnight. None of that was three pounds of fat burned in one night. It was the delayed appearance of fat loss that had been hidden by water.
Other common causes of temporary water retention include high-sodium meals, alcohol, intense exercise (which triggers an inflammatory response that holds fluid in muscles), travel (especially air travel), and certain medications like anti-inflammatories and some blood pressure drugs. Any of these can temporarily mask or amplify what the scale shows, and none of them reflect changes in body fat.
Why Body Composition Tools Struggle During Early Weight Loss
If you’ve ever stepped on a body-fat scale during the first weeks of a diet and gotten a bizarre reading, the explanation is rooted in water shifts. Most consumer-grade body composition tools, like bioelectrical impedance scales, work by sending a small electrical current through your body and measuring resistance. Water conducts electricity well, so more water means less resistance, which the device interprets as more lean mass. When you flush a lot of water during early dieting, the device may conclude you’ve lost lean mass even if your muscles are essentially unchanged.
Research on body composition after weight loss has found that the water fraction of fat-free mass can shift significantly. One study observed that the water fraction of the fat-free body increased after weight reduction (from about 73% to nearly 76%), which threw off several measurement methods and caused them to underestimate how much fat had actually been lost.12Metabolism. Assessment of fat-mass loss during weight reduction in obese women The bias was substantial, with some methods underreporting fat loss by nearly three kilograms.
More advanced techniques like multi-frequency bioelectrical impedance analysis and quantitative magnetic resonance can separately estimate total body water, extracellular water, fat, and lean tissue, giving a much clearer picture of what’s actually happening.13PubMed Central. Current body composition measurement techniques But these tools are typically available only in research settings or specialized clinics. For most people tracking weight at home, the practical takeaway is to ignore the body-fat percentage readout from a bathroom scale during the first month of a diet. The total weight number is at least consistent (you really did lose that weight), even if the composition breakdown the scale guesses is unreliable during a period of rapid water flux.
When the Water Comes Back
One of the most psychologically damaging moments in any diet happens when someone takes a break, eats normally for a day or two, and sees several pounds reappear overnight. This is almost entirely water. When you reintroduce carbohydrates, your body rebuilds glycogen stores and pulls water back in along with them, at the same three-to-four-grams-of-water-per-gram-of-glycogen ratio.2PubMed. Glycogen storage: illusions of easy weight loss, excessive weight regain, and distortions in estimates of body composition If you also eat saltier food, additional water follows sodium back into your tissues.
This rebound effect does not mean you regained fat. A person who dieted for three weeks, lost 10 pounds, then ate normally for a weekend and saw four pounds return has still lost roughly six pounds of actual body mass (fat plus some protein). The four pounds that returned was water and glycogen restocking. The problem is that the scale can’t tell you this, and the emotional impact of watching pounds “come back” drives many people to give up.
Understanding this pattern reframes the whole experience. The initial dramatic loss and the partial rebound after a diet break are both water phenomena, and they roughly mirror each other. Neither one tells you much about fat loss. The more reliable indicator is the trend over several weeks, ideally tracked as a moving average rather than daily weigh-ins.
The Motivational Paradox of Water Weight
Here’s where things get interesting from a behavioral standpoint. Even though early water loss isn’t “real” weight loss in the fat-burning sense, it may serve a genuine purpose. Research on people following very-low-energy diets has found that experiencing rapid initial weight loss helped sustain motivation. Participants described how seeing the scale move quickly reinforced their commitment and made the discomfort of calorie restriction feel worthwhile.14PubMed Central. Acceptability of a very-low-energy diet in Type 2 diabetes: patient experiences and behaviour regulation
This creates an awkward tradeoff. Diets that produce the most water loss early (very low-carb, very low-calorie) give the strongest motivational boost from the scale, but also set up the steepest apparent slowdown in weeks two through four as the body transitions to fat-dominant loss. Diets with moderate carbohydrate and calorie restriction produce a less thrilling first week, but the rate of loss stays more consistent, which may be easier to interpret correctly over time. Neither approach is inherently better for long-term fat loss. The “best” diet, as nutritionists have been saying for decades, is the one you can actually stick with.
Practical Ways to Tell Water From Fat
You can’t perfectly separate water loss from fat loss at home, but a few signals help you make educated guesses:
- Speed of change: If you lost more than two pounds in a single day, that’s water. Fat loss at even aggressive calorie deficits tops out at roughly half a pound of fat per day for most people, and typical deficits produce far less.
- Diet context: If you just started a low-carb diet or returned from a holiday of heavy eating and drinking, expect most of the first three to five pounds to be water. If you’ve been consistently dieting for six weeks, the loss you’re seeing is mostly fat.
- Physical signs: Frequent urination in the first few days of a new diet or fast is a hallmark of water shedding. If you’re going to the bathroom noticeably more often than usual, your body is dumping fluid.
- Rebound test: Eat a normal meal with carbohydrates and salt after several days of strict dieting. If two to four pounds reappear the next morning, that confirms the earlier loss was largely water. Your actual fat loss is approximately the net change once the rebound settles.
Waist measurements can also be more informative than the scale during the early weeks. Water distributes somewhat evenly throughout the body, while visceral fat concentrates around the midsection. A shrinking waistline alongside a flat scale is a sign that you’re losing fat but retaining or restoring water, which is a perfectly fine outcome even if the scale doesn’t reward it.