Story
The Number on the Body-Fat Scale
Last updated
In one pass You step on, and it reads 22.4%.
Educational content, not medical advice — consult a clinician.
Story path
Chapter 1
It does not measure fat
You step on, and it reads 22.4%. One decimal place: it looks like a measurement.
It is not a measurement. It is an inference.
A body-fat scale does exactly one thing. It passes a very small electric current through your feet (and your hands too, on models with handles) and measures how much resistance the current meets on its way through your body. That resistance is called impedance.
What it measures is not fat but how easily electricity travels. And how easily electricity travels through the body depends mainly on water: muscle, blood and organs hold a lot of water and conduct well; fat holds little water and conducts poorly. So the raw data the scale actually gets is a number closely tied to how much water is in your body.
Between that number and the 22.4% on the screen lie two more steps of inference — and fat itself is never measured, not even once.
It is not a measurement. It is an inference.
A body-fat scale does exactly one thing. It passes a very small electric current through your feet (and your hands too, on models with handles) and measures how much resistance the current meets on its way through your body. That resistance is called impedance.
What it measures is not fat but how easily electricity travels. And how easily electricity travels through the body depends mainly on water: muscle, blood and organs hold a lot of water and conduct well; fat holds little water and conducts poorly. So the raw data the scale actually gets is a number closely tied to how much water is in your body.
Between that number and the 22.4% on the screen lie two more steps of inference — and fat itself is never measured, not even once.
Mechanism · Why fat conducts poorly
Current does not travel through the body on electrons. It travels on ions: sodium, potassium, chloride — charged particles dissolved in water. Wherever there are more ions and more water, there is an easier path.The inside of a fat cell is almost entirely taken up by one large droplet of , with very little water or ions. For a current, that is a stretch of road with almost no charge carriers. Muscle, blood and organs are the opposite extreme: high in water, with stable electrolyte levels, they are the body's natural wiring.
So high impedance roughly means less water-holding tissue. The relationship is real and has a physical basis; it is not marketing, and the whole design of a body-fat scale rests on it (the Kyle 2004 methods review describes exactly these principles).
The problem is not this step. The problem is that getting from how much water-holding tissue there is to how much fat there is takes two more steps, and each carries an assumption that may not hold for you. The chapter Fat is the leftover takes that chain apart.
Chapter 2
Fat is the leftover
From impedance to that percentage runs a three-step chain of inference, and each step rests on an assumption.
Step one, from impedance to total body water. It assumes the body is a conductor of even thickness. It is not: the limbs are long and thin, and almost all the resistance sits there; the trunk is short and thick, carries most of your weight, and offers almost no resistance. A current from one foot to the other is mostly measuring your legs.
Step two, from total water to fat-free mass (all of your weight that is not fat). It assumes that the share of water in fat-free mass is a fixed value (about 73%). That share differs from person to person and drifts with your condition.
Step three, from fat-free mass to fat. This step is subtraction: body weight minus fat-free mass, and whatever is left is called fat.
Step three is the crux. Fat is never measured at any point; it is the remainder, and every error from the first two steps ends up landing on it.
Step one, from impedance to total body water. It assumes the body is a conductor of even thickness. It is not: the limbs are long and thin, and almost all the resistance sits there; the trunk is short and thick, carries most of your weight, and offers almost no resistance. A current from one foot to the other is mostly measuring your legs.
Step two, from total water to fat-free mass (all of your weight that is not fat). It assumes that the share of water in fat-free mass is a fixed value (about 73%). That share differs from person to person and drifts with your condition.
Step three, from fat-free mass to fat. This step is subtraction: body weight minus fat-free mass, and whatever is left is called fat.
Step three is the crux. Fat is never measured at any point; it is the remainder, and every error from the first two steps ends up landing on it.
Mechanism · Whose 73% is it
The constant in step two deserves a look of its own, because it is the quietest assumption in the whole chain.It is a population average. Fat-free mass is not one uniform substance. It is a mixture of muscle, bone, organs, blood and connective tissue, and each of these holds a very different amount of water. If the proportions of these components change in a given person, the overall water share changes with them.
The standard methodological reference for bioelectrical impedance analysis (BIA) — Kyle 2004, from the working group of the European Society for Clinical Nutrition and Metabolism (ESPEN) — puts it more firmly: BIA can determine fat-free mass and total body water only in people without significant fluid and electrolyte abnormalities, and only when the equation used is specific to that population, age or disease state. In other words, a home scale that applies one equation to everyone is, methodologically, already outside those conditions.
Who drifts furthest from the average:
People with very high or very low muscle mass: off in both directionsOlder adults: body composition shifts with agePeople with edema or fluid retention: more water, but that is not more fat-free massChildren and adolescents: the water share of a growing body differs from an adult's to begin with
Follow the chain and you reach a counterintuitive result: someone with edema has more body water, so the estimated fat-free mass comes out higher, and after the subtraction, fat comes out lower. The scale will tell you your body fat has dropped because you are swollen.
That is why "step three is subtraction" matters so much: it turns every upstream error into a wrong conclusion about fat, with nothing on the display to warn you.
Chapter 3
How far off is it
So how far off does this chain end up?
To answer that, you first need a more accurate ruler. Research uses the four-compartment model: it measures body weight, body volume, body water and bone mineral all at once, splitting the body into four parts, and it is currently accepted as the approach closest to the true value.
A 2026 systematic review pooled 12 validation studies in healthy adults that used the four-compartment model as the standard, and it gives two numbers:
Mean bias: −3.5% to +4.4% (some devices run systematically high, some low)Limits of agreement: typically spanning 15 to 20 percentage points
The second number is the crux. It means that when the scale reads 25%, your true body fat lies somewhere between roughly 17% and 33%.
That is not a body-fat percentage. It is a range, and one wide enough to cross several categories.
To answer that, you first need a more accurate ruler. Research uses the four-compartment model: it measures body weight, body volume, body water and bone mineral all at once, splitting the body into four parts, and it is currently accepted as the approach closest to the true value.
A 2026 systematic review pooled 12 validation studies in healthy adults that used the four-compartment model as the standard, and it gives two numbers:
Mean bias: −3.5% to +4.4% (some devices run systematically high, some low)Limits of agreement: typically spanning 15 to 20 percentage points
The second number is the crux. It means that when the scale reads 25%, your true body fat lies somewhere between roughly 17% and 33%.
That is not a body-fat percentage. It is a range, and one wide enough to cross several categories.
Numbers · Bias and limits of agreement differ
These two numbers are often used interchangeably, but they answer completely different questions, and only by telling them apart do you learn what the scale can do.Mean bias answers: does this device run high or low overall? It is the average error across a group of people. If a device has a bias of +3%, it overestimates body fat by 3 percentage points on average. That part is systematic: for the same person, it pushes the reading in roughly the same direction every time.
Limits of agreement answer: for one particular person, how far off can a single reading be? They describe spread, not the average.
Why the second matters more: you are not a group; you are one person standing on a scale wanting your own number. Mean bias can in principle be subtracted afterward; spread cannot.
Kyle 2004 supplies the other half of the reason from the methods side: BIA is widely used but has long lacked a standardized method and quality-control procedures. This is not one manufacturer cutting corners; the method as a class has never been standardized.
One practical conclusion follows directly: the review stresses that performance differs widely between devices, and agreement data from one analyzer cannot be carried over to another. So a reading from the gym machine, one from your home scale and one from a health check are not comparable — they are not three marks on one ruler but three different rulers.
Chapter 4
Wrong, but not for that reason
Once people learn that it is inaccurate, most ask next: if I skip water, food and exercise before measuring, will it be accurate then?
Here the result runs against intuition. A 2021 study set out to deliberately break those preparation rules: 40 people and 3 home body-fat scales, measured while dehydrated, after exercise, right after drinking water, right after a meal and with a full bladder, each compared with readings taken under standard conditions.
The result: no condition differed significantly from the control. 86% of the tests fell within 2 percentage points of the control value, and 97% within 5. The errors were largest in the dehydrated condition, and women were more likely than men to exceed 2 points.
In other words, it really is inaccurate, but not because you drank that glass of water. The problem lies in the chain of inference from impedance to fat itself, not in the two hours before you step on.
Here the result runs against intuition. A 2021 study set out to deliberately break those preparation rules: 40 people and 3 home body-fat scales, measured while dehydrated, after exercise, right after drinking water, right after a meal and with a full bladder, each compared with readings taken under standard conditions.
The result: no condition differed significantly from the control. 86% of the tests fell within 2 percentage points of the control value, and 97% within 5. The errors were largest in the dehydrated condition, and women were more likely than men to exceed 2 points.
In other words, it really is inaccurate, but not because you drank that glass of water. The problem lies in the chain of inference from impedance to fat itself, not in the two hours before you step on.
Evidence · What this study did and did not show
This study is unusually easy to read backwards, so here are its boundaries. It compared different conditions within the same group of people; it did not check the readings against a more accurate method.What it shows: breaking the preparation rules over a short window barely moves the reading. That is, the reproducibility of these devices is better than gym lore assumes.
What it does not show: that the reading is correct. Reproducibility and accuracy are two different things. A ruler with misprinted markings gives you the same wrong number every time you measure the same table: it is extremely stable, and it is always wrong.
Accuracy is measured by a different study: the 2026 systematic review that compared these devices with the four-compartment model, where the limits of agreement span 15-20 percentage points. Put the two together and the picture is stable but offset — the most deceptive kind of error there is, because stable looks so much like accurate.
One detail should not get lost: the largest errors did come from dehydration. So the manual's line about not measuring while dehydrated is not entirely wrong. It has simply been inflated into a whole ritual, and the ritual does not touch the real problem.
This is also why you should watch the trend: a stable offset partly cancels out when you subtract one reading from another, whereas the error in the absolute value stays, however carefully you prepare.
Chapter 5
So how should you use it
So the conclusion is not to throw the scale away but to ask it only the questions it can answer.
What it cannot answer: what is my body-fat percentage right now? The uncertainty around that absolute value is too wide to tell you which category you are in.
What it can answer: has the direction changed? With the same device, at the same time of day and in the same condition, readings over several weeks carry roughly the same systematic offset each time, so part of it cancels out when you subtract one from another. The trend is far more trustworthy than the absolute value.
A better use is to demote it to one signal among several:
Waist circumference: one soft tape measure, cheap and repeatable; it reflects fat around the abdomen and is closer to visceral fat than a whole-body fat percentagePhotos taken under the same conditionsStrength and girth: a bench press that keeps climbing while your waist shrinks tells you far more than 22.4% becoming 21.8%
What it cannot answer: what is my body-fat percentage right now? The uncertainty around that absolute value is too wide to tell you which category you are in.
What it can answer: has the direction changed? With the same device, at the same time of day and in the same condition, readings over several weeks carry roughly the same systematic offset each time, so part of it cancels out when you subtract one from another. The trend is far more trustworthy than the absolute value.
A better use is to demote it to one signal among several:
Waist circumference: one soft tape measure, cheap and repeatable; it reflects fat around the abdomen and is closer to visceral fat than a whole-body fat percentagePhotos taken under the same conditionsStrength and girth: a bench press that keeps climbing while your waist shrinks tells you far more than 22.4% becoming 21.8%
In practice · Where the more accurate methods sit
If you really do need an absolute value that is as accurate as possible, here is where each method stands. But the conclusion first: most people do not need one, because the decisions in front of you (what to eat, how to train, how long to keep going) do not change whether that number is 22% or 26%.Four-compartment model: the reference standard in research, which requires several measurements at once. It exists essentially only in laboratories; it is not a service you can bookDual-energy X-ray absorptiometry (): common in clinics and research, and it also reports where fat is distributed and your . It is generally more reliable than a body-fat scale, but it has its own error and needs dedicated equipment and a low-dose X-rayVolume-based methods (BodPod) and underwater weighing: these measure body density and infer composition from it. Note that this, too, is an inference, just one that rests on fewer assumptionsSkinfold thickness: cheap, but heavily dependent on the operator's technique; the same person measured by someone else gets a different result
Among the reference methods the systematic review itself recommends are more expensive imaging methods such as magnetic resonance imaging (MRI).
Reading down that list, one thing becomes clear: every method is indirect. The only way to measure body fat directly is dissection, so there is no true value to be had here, only inferences with larger or smaller errors.
That is not nihilism; it is the point of this story. Knowing how a number was produced is what tells you how much conclusion it can bear. What a body-fat scale can bear is a trend, not a verdict.
References · 3
- Kyle, U. G., Bosaeus, I., De Lorenzo, A. D., Deurenberg, P., Elia, M., Gomez, J. M., Heitmann, B. L., Kent-Smith, L., Melchior, J. C., Pirlich, M., Scharfetter, H., Schols, A. M., & Pichard, C. (2004). Bioelectrical impedance analysis - part I: review of principles and methods. Clinical Nutrition, 23(5), 1226-1243. ESPEN Working Group methodological review. States that BIA use is widespread but suffers from a lack of standardised method and quality-control procedures, and that BIA allows determination of fat-free mass and total body water only in subjects WITHOUT significant fluid and electrolyte abnormalities, and only when using population-, age- or pathology-specific BIA equations together with established procedures. It also states that the use of segmental BIA, multifrequency BIA or bioelectrical spectroscopy in altered hydration states requires further research. 10.1016/j.clnu.2004.06.004
- Oliver, C. J., Del Vecchio, L., Minehan, M., Climstein, M., Rosic, N., Myers, S., & Tinsley, G. (2026). The validity of bioelectrical impedance analysis compared to a four-compartment model in healthy adults: A systematic review. Journal of Functional Morphology and Kinesiology, 11(1), 65. Twelve validation studies in healthy adults against the four-compartment criterion model. Mean bias for percentage body fat ranged from -3.5% to +4.4%, and limits of agreement typically spanned 15 to 20 percentage points. Conclusion: BIA estimates were overall not equivalent to the 4C model, performance is device-specific so agreement from one analyser must not be generalised to another, and alternative criterion methods such as MRI are recommended. 10.3390/jfmk11010065
- Randhawa, A. K., Jamnik, V., Fung, M. D. T., Fogel, A. S., & Kuk, J. L. (2021). No differences in the body fat after violating core bioelectrical impedance measurement assumptions. BMC Public Health, 21(1), 495. Within-subject study, n=40 across three consumer BIA devices, measuring percent fat mass under control, dehydration, exercise, water intake, meal intake and non-voided bladder conditions. No condition differed significantly from control, range -1.9 to 0.4 percent; 86 percent of tests fell within 2 percentage points of control and 97 percent within 5, despite deliberately violating the manufacturer preparation rules. Errors were largest with dehydration, and women were more likely than men to exceed a 2-point difference. 10.1186/s12889-021-10552-y