I'm a bit overweight. I acquired some home body impedance scales (not entirely sure how accurate these are) and the app has ranges on it for normal, overweight, etc.

Specifically, currently I am 107.6kg, 29.0% fat, 45.9% muscle. I've been looking at my targets. For my height, the top of the "normal" boundary is 83.72kg, and 25% fat.

Here's the problem. My current fat weight in is 31.2kg. My fat weight with the "normal" range targets is 20.93kg- so I'm trying to lose 10.3kg of fat. But the total weight difference between my total current weight and my targetted normal weight is, uh, 24kg. If I actually lost 24kg of fat, I would be at just 6kg of fat which is insane for a 84kg person and far from the "normal" upper end which is 21kg.

So if I'm trying to lose weight down to 84kg, what the hell am I actually losing? It seems like if I just lose 10.3kg of fat, I'll still be way over "normal" weight.

I also noticed that if I keep the same muscle mass I'll be 59% muscle at 84kg which is entering the "high" end. Since I'm pretty unfit (although getting slightly fitter) it seems like this can't be right.

  • 1
    Those scale readings are pretty suspicious. Could you provide your age, sex and height to help interpreting those percentages? Commented May 18, 2020 at 10:01
  • I'm male, 29, and 1.83m. What's suspicious about them?
    – Puppy
    Commented May 18, 2020 at 10:24
  • 1
    The suspicious part would be how your FFMI of nearly 23 coincides with you being pretty unfit. FFMI is like BMI for your lean (non-fatty) weight. It’s essentially a measure of muscularity which is computed by having a point of data from body fat percentage along with the standard height and weight. Specifically your FFMI of 22.63 sounds wrong for two reasons. The first was already mentioned (the discrepancy between your description and the given number) while the second is that those scales (especially the cheaper ones) tend to have a large margin of error. Commented May 18, 2020 at 11:35
  • 46% muscle mass is extraordinarily high and doesn't really check out with being overweight and 'pretty unfit'. Also, labelling 25% body fat as "normal" is questionable, imo. Most institutions such as the American Institute on Excercise and the NIH rate 25% in your demographic as "obesity". Have you looked at the various bodyfat comparison pictures on the internet? duckduckgo.com/… Do your measured 29% roughly check out with what is depicted there, or does it seem off by a lot, say 10% or more? Commented May 18, 2020 at 11:46
  • It's more the 30-34% range. I guess that the bottom line then is that the scales and values listed in the app aren't all that accurate.
    – Puppy
    Commented May 18, 2020 at 12:02

2 Answers 2


The goal of weight loss is almost always fat loss despite the phrases being used interchangeably by the general public. When you lose weight it isn't always clear what you are losing, but there are some things that you can keep in mind in order to better understand what's going on.

The "3500 calorie per pound of fat" model of weight loss isn't perfect and has some issues, but it's close enough to be useful. When your body is in an energy deficit, it releases stored energy to balance out that deficit. So if your body is 500 calories below maintenance on a particular day, your body will release energy from its stores equal to 500 calories to balance that out. Your body will not seek to use more or less than it needs, so a deficit of 500 calories always results in 500 calories worth of energy being broken down and released. So what is stored energy? Fat, glycogen, and muscle. Fat is the body's preferred method of energy storage, glycogen (carb storage) is meant to be used for immediate energy (running, lifting weights) as opposed to chronic maintenance like fat, and muscle (protein storage) isn't meant to be used for energy storage at all, but it will be used as such if the circumstances are right. How much is each of these worth?

  • Fat contains 3500 calories per pound (2.2 kg).
  • Glycogen contains ~1800 calories per pound.
  • Skeletal Muscle contains ~600-700 calories per pound.
  • Water contains ZERO calories per pound.

It's important to note that water fluctuations can and will often mask energy balance on the scale. This can occur by simply drinking more or less liquids than normal, but it can also occur by manipulating glycogen stores. Every ONE gram of glycogen requires THREE grams of water to store it. So while one pound of glycogen may be worth 1800 calories, losing that one pound of glycogen also means losing three additional pounds of water for a total of four pounds being lost. The first week of a diet (up to the first 2-3 weeks) is typically marked by huge shifts in water and glycogen stores in the body. Not only that, but the weight of the food sitting in your digestive system typically goes down as well. All this results in the initial 5-10 pounds of weight loss being short term adaptations as opposed to long term adaptations in weight. Another good rule of thumb is that for every five pounds of weight that you lose, one pound of that will be water (or put another way, 20% of your weight loss will typically be water).

What about muscle mass? How do you prevent the breakdown of muscle tissue? You give it what it needs. Exercise tells your body that the muscles being used are important and protein helps to maintain and repair muscle mass. You can actually lose weight faster if don't do any exercise at all, but that's because more of the weight being lost is coming from muscle. If you lost weight entirely from muscle mass, a daily deficit of 500 calories (weekly deficit of 3500) would result in nearly 6 pounds of weight loss! But that would carry a heavy negative impact on body composition and your capacity to do daily activities. The reality is that weight loss always comes from multiple sources and muscle loss can be minimized or even reversed by simply exercising appropriately.

So what's a reasonable expectation for the composition of weight loss? After the first three weeks of dieting where glycogen stores as well as water and food weight have stabilized, I would break it down as the following ESTIMATION for anyone above 15% bodyfat who is doing resistance training and feeding their body appropriately.

  • 20% Water
  • 80% Fat

Finally, if you are gaining or losing weight much more rapidly than your calorie intake suggests that you should be, you can reasonably assume that it is water/glycogen. Unless you just went on a binge-fest all week. The best way to monitor this is by weighing daily first thing in the morning when you wake up from sleep. Doing this will allow you to take weekly means or medians which will be a good representative of your non-fluctuating weight each week as opposed to weighing once a week and hoping that it isn't higher or lower than it should be reporting. There isn't any benefit to weighing more than daily, but if that seems like too much, twice a week could also work. Personally, I just take a quick picture of the scale and write down the information later in a chart.

  • +1 This is a great overview of the truth of what people see when the scale is changing.
    – michael
    Commented May 18, 2020 at 17:09

I can only comment on the impedance measurements. I recently (3 years) started using them and have found huge variation in spot measurements. The trick is to measure consistently.

  1. Usually the same time of the day. Morning or evening. Also make sure you poop before a measurement. Do not measure after exercising.
  2. Measure regularly. Not every day, but use the average of two or three weeks.
  3. Take it as relative. Don't quote the figures as 21kg -> 10kg fat, Rather say I have made a (quantitatively measurable and statistically significant) improvement over the last 8 weeks.

Those machines aren't exact but when used (skeptically, patiently, and correctly) are helpful to quantify your progress.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.