5.2 pounds. That was the difference between my weight at 10 PM on Thursday and my weight at 6 AM on Friday. I did not eat 18,200 calories in my sleep. I did not run a marathon in my dreams. I just slept. And my weight swung by 5.2 pounds. If I had only weighed myself on Thursday evening, I would have panicked. If I had only weighed myself on Friday morning, I would have celebrated. Both numbers were real. Both numbers were meaningless. And the story of why your weight swings overnight is the story of why daily weigh-ins are a terrible way to track health.
I ran an experiment. For 72 hours, I weighed myself every hour. Yes, every hour. For three days. I am a data guy. I used to build dashboards for stock prices. Now I build them for my own body. And I wanted to see the variance. The noise. The chaos that lives inside the number we treat as gospel. The result was a graph that looked like a stock market chart during a volatile earnings week. Peaks and valleys. Spikes and crashes. And not a single number that told the truth about my body.
Here is what the data says. I used a calibrated digital scale. Same scale. Same location. Same conditions. I weighed myself after bathroom use but before eating or drinking. Even with these controls, the hourly variance was shocking. Over 72 hours, my weight ranged from 179.8 lbs to 185.0 lbs. A 5.2-pound spread. The coefficient of variation was 0.72%. That is enormous for a biological measurement. For context, my resting heart rate varies by 3% day-to-day. My blood pressure varies by 5%. My weight varies by 3%. But within a single day, it varies by 3% from morning to night. The daily swing swallows the weekly trend.
Calculate your BMI instantly with visual category charts and history storage.
All data stays in your browser — we never see it.The drivers of the swing are well-understood but poorly appreciated. Water is the biggest factor. The human body is 60% water. A 180-pound person carries 108 pounds of water. A 1% change in hydration status is 1.8 pounds. A 2% change is 3.6 pounds. And 2% hydration changes happen constantly. You wake up dehydrated after 8 hours without water. You drink 16 ounces of coffee. You eat a salty lunch. Your body retains sodium, which retains water. You exercise and sweat. You drink water to rehydrate. Each of these events shifts your water balance by 1-3 pounds. And the scale cannot tell water from fat.
Glycogen is the second factor. The body stores carbohydrates as glycogen in the liver and muscles. Each gram of glycogen binds 3-4 grams of water. A typical adult stores 400-500 grams of glycogen. That is 1,200-2,000 grams of water — 2.6-4.4 pounds. When you eat a low-carb meal, your body depletes glycogen and releases the bound water. Weight drops 2-3 pounds overnight. Not fat loss. Water loss. When you eat a high-carb meal, your body replenishes glycogen and binds water. Weight jumps 2-3 pounds overnight. Not fat gain. Water gain. The scale is measuring your dinner, not your body.
Sodium is the third factor. A high-sodium meal increases extracellular fluid volume through osmotic retention. A single salty dinner — say, BBQ from La Barbecue on East Cesar Chavez — can increase water retention by 2-3 pounds for 24-48 hours. The sodium is eventually excreted by the kidneys. But the temporary weight spike is real. And it is meaningless. I tested this. Day 1: normal sodium, 2,300 mg. Morning weight: 181.2. Day 2: high sodium, 5,800 mg (tacos, chips, salsa). Morning weight: 183.7. Day 3: normal sodium. Morning weight: 181.4. The 2.5-pound swing was entirely sodium-driven water retention. No fat was gained or lost.
Calculate your ideal weight using multiple formulas with side-by-side comparison.
All data stays in your browser — we never see it.Fiber is the fourth factor. Undigested food in the gastrointestinal tract adds weight. A high-fiber meal can increase gut content by 1-2 pounds. This is not body weight. It is food weight. But the scale cannot distinguish. Constipation can add 2-3 pounds of stool weight. Laxation can remove it. The scale sees both as body weight changes. They are not.
Hormones are the fifth factor. In women, the menstrual cycle causes water retention fluctuations of 2-5 pounds. I do not have this variable. But I tracked cortisol indirectly through HRV. High stress days — deadline pressure, Pixel's vet visit, an argument with my ex-wife about whose turn it was to pick up the dog — correlated with morning weight increases of 0.5-1.5 pounds. Cortisol increases sodium retention and shifts fluid distribution. The stress weight is real. And it is not fat.
I modeled the components of the 5.2-pound swing. Water: 3.1 pounds. Glycogen: 1.2 pounds. Sodium: 0.6 pounds. Fiber/gut content: 0.3 pounds. Total: 5.2 pounds. Fat change: zero. Muscle change: zero. The entire swing was non-fat mass. The scale was measuring my hydration, my dinner, my stress, and my bowel movements. It was not measuring my body composition. It was measuring my lifestyle in the previous 24 hours.
Estimate body fat percentage using the US Navy tape method and BMI-based formulas.
All data stays in your browser — we never see it.The solution is the moving average. Not the daily weight. Not the weekly weight. The 7-day or 14-day moving average. I calculated this from my 72-hour data. The daily weights were: 181.2, 183.7, 181.4, 182.1, 180.8, 181.9, 180.5. The 7-day moving average: 181.7. The variance of the moving average was 0.3 pounds. The variance of the daily weights was 2.8 pounds. The moving average reduced noise by 89%. It was not perfect. But it was usable. It was honest. It was the signal extracted from the noise.
I also tested a "trend weight" method. Weigh daily. Take the median of the last 7 days. Compare to the median of the previous 7 days. If the current median is lower than the previous median by more than 0.5 pounds, you are losing weight. If it is higher by more than 0.5 pounds, you are gaining. If it is within 0.5 pounds, you are stable. This method is robust to daily swings. It requires daily weighing — which some people find anxiety-inducing — but it protects against the false signals of single measurements.
I brought this data to a Gevity meetup on East Cesar Chavez. A woman named Rebecca — she is a registered dietitian — said something that stuck with me. "I tell my clients to throw away their scale or use it as a paperweight. The daily number is not data. It is trauma." She was not wrong. The psychological impact of daily weigh-ins is well-documented. People who weigh daily have higher rates of body dissatisfaction, disordered eating, and anxiety. The number on the scale becomes a moral judgment. Good day, bad day. Success, failure. When the number is actually just water.
But I am a data guy. I do not throw away data. I process it. I filter it. I extract signal from noise. So here is my compromise. Weigh daily if you want the data. But do not look at the daily number. Look at the 14-day moving average. Look at the trend. Look at the direction, not the snapshot. And never, ever make a decision based on a single reading. Never skip a meal because the scale was up 2 pounds. Never celebrate because the scale was down 3 pounds. The 2 pounds is water. The 3 pounds is glycogen. The real change happens over weeks and months. And it is measured in fractions of pounds, not whole pounds.
I also recommend controlling the measurement conditions. Same time of day — morning, after bathroom, before coffee. Same scale — calibrated, on a hard surface. Same clothing — ideally none. Same hydration status — drink a standard amount of water the night before. These controls reduce variance by 30-40%. They do not eliminate it. But they make the signal clearer. And clearer signal means better decisions. Better decisions mean better health. Even if the health is just mental — less anxiety, less obsession, less trauma from a number that does not mean what we think it means.
So here is my challenge. Try the moving average method for 30 days. Weigh every morning. Log the number. Do not react to it. At the end of 30 days, calculate the 7-day moving average for the last week and compare it to the first week. That is your trend. That is your signal. The daily numbers are noise. The trend is truth. And the truth is usually boring. Boring is good. Boring means stable. Boring means sustainable. Boring means you are not riding the emotional roller coaster of daily weight swings.
I will keep the spreadsheet open. I will keep weighing hourly when I am feeling particularly obsessive. I will keep calculating moving averages. And I will keep reminding myself — and you — that 5.2 pounds overnight is not a body change. It is a water change. It is a glycogen change. It is a sodium change. It is a stress change. It is everything except fat. And fat is what we care about. Fat is what determines health. Fat is what the scale cannot measure. And that is why we need better tools. Better metrics. Better understanding. And better emotional regulation when the number goes up. Because it will go up. And it will go down. And neither direction means what we think it means. The data is clear. The scale is a liar. But it is an honest liar. It tells you exactly what you weigh. It just does not tell you what that weight means. And that is your job. Not the scale's. Yours. And mine. And Pixel's, who does not care about any of this and just wants her walk. Which is probably the healthiest perspective of all.
