Skinny fat sounds like an insult invented on the internet, which it partly was. It also describes something measurable, and the measurement predates the slang by a good margin.
The short answer
Skinny fat is the informal name for normal weight obesity: a body mass index that sits inside the normal range, between 18.5 and 24.9, paired with a proportion of body fat that does not. You can weigh what a chart says you should weigh and still carry a body composition that behaves, metabolically, like a heavier person's.
The clearest data comes from a 2010 analysis in the European Heart Journal of 6,171 adults from NHANES III, all of whom had a normal BMI and a full body composition assessment.
The term researchers actually use
Normal weight obesity is the phrase in the literature. The researchers divided their normal BMI sample into sex specific tertiles of body fat percentage and labelled the top third as normal weight obesity. In that sample the cut fell above 23.1% body fat for men and above 33.3% for women.
Those are not universal thresholds, and it matters that they came out of one dataset. But they give the term a numerical shape, which is more than the slang version has.
What 6,171 adults showed
Everyone in the sample had a normal BMI. The differences came entirely from what the body was made of.
The prevalence of metabolic syndrome in the highest body fat group was 16.6%, against 4.8% in the lowest, which is roughly a four fold difference at the same body weight category. The high body fat group also had higher rates of dyslipidaemia, higher rates of hypertension among men, and higher rates of cardiovascular disease among women.
Survival data was available for more than 99% of the sample over a median follow up of 8.8 years. After adjustment, women in the normal weight obesity group had a 2.2 fold increased risk of cardiovascular mortality compared with the low body fat group. The equivalent association did not reach significance in men in this analysis.
Larger cohorts have found related patterns. A 2021 prospective study in Clinical Nutrition following 463,002 adults reported that adiposity measures tracked mortality risk in ways BMI alone did not fully capture.
Why BMI cannot detect it
BMI is height and weight. That is the whole calculation. It has no way to distinguish a kilogram of muscle from a kilogram of fat, because it never sees either.
The bathroom scale has the same blind spot for the same reason. Two people at identical height and weight can differ by fifteen percentage points of body fat and produce the same number on both instruments. This is not a flaw in how you are using them. It is what they measure.
It also explains why the pattern is so often invisible to the person in it. Nothing they check reports a problem, because nothing they check is looking at composition.
Why it shows up after weight loss
This is the version that matters for anyone reading this site. Weight lost quickly is not lost proportionally. Fat and lean tissue both come off, and the lean share of the loss rises when protein intake is low, when resistance training is absent, or when the deficit is steep.
Arrive at a normal BMI having shed a meaningful amount of muscle along the way and you can land squarely in the normal weight obesity category despite having done everything the scale asked. The number is correct. The composition underneath it moved in the wrong direction.
That is one of the specific failure modes this app was built around, and it is the reason the two features it leads with are a protein floor and short strength sessions rather than anything to do with eating less.
What actually changes the ratio
Only two things reliably move body composition at a roughly stable body weight, and neither is a diet.
The first is resistance training, which supplies the signal for the body to keep and build lean tissue. The second is eating enough protein for that signal to have material to work with. Cardio contributes to the fat side and does very little for the lean side, which is why cardio alone tends to produce a smaller version of the same ratio.
The frustrating part is that this process is slow and the scale is unhelpful throughout it. Adding lean tissue while losing fat can hold your weight completely flat for months while your body composition changes underneath. If the scale is your only instrument, that period reads as failure.
Why OffRamp does not estimate body fat
OffRamp deliberately does not calculate or display a body fat percentage. Consumer estimates from tape measurements or bioimpedance carry error margins wide enough that the number tells you very little, and showing one anyway would give a false precision to something the app cannot actually measure.
What it does instead is act on the two levers that change composition. The protein floor is set at 1.6 grams per kilogram of your goal weight, rounded to the nearest 5 grams, and it is a floor rather than a target because the common failure in maintenance is falling short of it. Three strength plans ship with the app, one bodyweight, one with dumbbells and one with bands, at two 20 minute sessions a week. The bodyweight session opens with three sets of twelve bodyweight squats and three sets of ten incline push ups, and the prescription is deliberately fixed rather than progressive, because two sessions completed beats four sessions planned.
Weight is still tracked, just not trusted on any single day. Regain Radar reads a seven day rolling average against your baseline and holds its tongue until that average sits at or above baseline times 1.02 for seven consecutive days. That gives a recomposition period room to be flat without the app treating flat as failure.
If your body composition is a medical concern, that is a conversation for a clinician who can measure it properly. OffRamp does not provide medical advice, and anything involving a prescribed treatment belongs with your prescriber.
