Body Type Preferences in Dating: The Data

Most people, when asked, say they want a partner in a healthy weight range, but the single adults they are actually choosing from are heavier than that preference assumes, and the mismatch creates a quiet supply gap. On the demand side, stated preferences cluster around an average or fit build. On the supply side, the measured reality is blunt: about 40 percent of US adults are obese and a further large share are overweight, so a majority of the population sits above the band that preference surveys tend to favor. That gap does not make a body-type filter useless, but it does mean the filter behaves differently than people expect. This page summarizes the general, well-established findings on stated body-type preferences, sets them against the measured body mass index (BMI) distribution of US adults, and explains why a body or fitness filter cuts more gently than a height or income floor.

What people say they want

Preference research on body type is messier than research on height or income, so the honest summary is a set of directions rather than precise figures. Surveys and dating-platform studies broadly agree on a few points. Most respondents, when they describe an ideal partner, gravitate toward words like average, fit, athletic, or in shape. Very few describe a preference for either extreme of the weight range. The middle of the distribution, in other words, is where stated demand concentrates.

Those findings come with heavy caveats, which is why any specific percentage should be treated with suspicion. People report ideals that are thinner than the partners they actually date and marry. Stated preferences shift with the wording of the question, the format of the survey, and whether the respondent is picturing a fling or a marriage. What someone selects on a checkbox is a poor guide to who they swipe on, and both are poor guides to who they end up with. So the safe reading is a shape, not a number: demand leans toward the healthy-weight and mildly athletic bands, and thins out as you move toward either tail.

How preferences split by gender

The gender pattern is one of the more stable findings in this area, though the size of the gap varies from study to study. On average, men place somewhat more weight on a partner's physical build and appearance than women do, and women place somewhat more weight on traits like height, resources, and status. This shows up across decades of preference research and across different methods, which is why it is treated as well documented rather than speculative.

In practice that means a body-type preference tends to bite a little harder when a man is doing the filtering and a little softer when a woman is. It does not mean women ignore build or that men ignore everything else. The differences are averages across large groups, with enormous overlap between individuals. Our guide on every filter explained and the stats page on whether women prefer tall men both cover how these gendered tendencies play out across the other inputs, where height carries more of the load on one side and build carries more on the other.

One more caution belongs here. Stated body-type preferences are the least reliable of the common dating filters precisely because they mix physical measurement with taste, health, and social judgment in ways that self-report captures badly. Height is a hard number. Income is a hard number. Build is partly a number and partly an impression, which is why the calculator treats it as a softer input.

The measured reality: BMI of single adults

Against that fuzzy demand sits a hard supply figure. CDC NHANES, the National Health and Nutrition Examination Survey, measures the height and weight of a representative sample of Americans directly rather than asking them to report it. From those measurements it estimates BMI and sorts the adult population into standard bands: underweight below 18.5, healthy weight from 18.5 to 24.9, overweight from 25 to 29.9, and obese at 30 and above.

The recent picture is consistent and heavy. About 40 percent of US adults fall in the obese band, a share that has risen over the past several decades. Roughly another 30 percent fall in the overweight band. Add those together and about 70 percent of US adults measure as overweight or obese, which leaves the healthy-weight band holding a minority, somewhere in the region of a quarter to a little over a quarter of adults. The underweight band is small, only a couple of percent. These are population-wide figures, and they shift somewhat by age and sex, but the headline holds across the adult range that matters for dating.

That is the number worth sitting with. If stated demand concentrates in the healthy-weight and mildly athletic bands, and only about a quarter of adults measure inside the healthy band, then the group most people say they prefer is a minority of the people actually available. The single population is not identical to the full adult population, since it skews a little younger, but not by enough to close the gap. For the wider supply picture, the stats page on the US dating pool lays out how many single adults there are to begin with.

Stated preference versus measured distribution

The table below places the two sides next to each other. The measured column comes from the CDC NHANES BMI distribution and is expressed in approximate terms, since the exact share moves a little with each survey wave. The preference column is deliberately qualitative, because the honest state of the research does not support precise percentages for what people say they want. Read it as direction, not measurement.

Two things jump out of the table. First, the single largest band, obese at about 40 percent, is the one preference surveys favor least, while the band people favor most, healthy weight, is smaller than either of the two heavier bands. Second, the overweight band is large and sits in an ambiguous place: many people accept a partner there in practice even when they would not name it as their stated ideal. That gap between what people accept and what they claim to want is exactly where the supply mismatch lives.

Why this creates a supply gap

A supply gap appears whenever demand concentrates on a slice of the population that supply cannot fill. Height is the classic example: many people want a tall partner, but only about 1 in 7 men reach 6 feet, so the demand outruns the supply and the filter cuts hard. Body type produces a milder version of the same effect. Demand leans toward the healthy-weight and athletic end, but that end holds only about a quarter of adults, so anyone insisting on it is fishing in a smaller pond than they picture.

The difference is that the body-type gap is more forgiving in two ways. The healthy-weight and overweight bands together still cover more than half the population, so a preference that stretches to include the lower overweight band, which many people do in practice, keeps a wide pool. And build is changeable in a way height is not, so the boundary is soft at the edges. Both of these keep the body-type filter from biting as sharply as its raw supply figure might suggest.

How a body-type filter stacks with other filters

Inside the calculator, body type is one of six filters, alongside age range, height, income, education, and marital status. Each filter keeps only the share of the population that meets your setting, and the filters compound as you add them. A body-type filter on its own removes a modest slice, usually the heaviest tail of the distribution, which might be a third of the pool or less at an average setting. That is a real cut, but a gentle one next to a height or income floor.

The interaction matters more than the standalone number. Body, height, and income do not vary independently in the real population, they correlate, which changes how the cuts stack. Because these traits cluster, a person who already clears a height and income bar is somewhat more likely to clear an average-build bar too, so adding body type on top of those filters removes less than a naive multiplication would predict. Multiplying the raw frequencies would overstate how rare the combination is. The calculator applies a correlation adjustment instead, which the how it works guide walks through step by step. The short version: a body-type filter layered onto a strict height and income wish list adds a smaller marginal cut than it would to a blank slate.

Why body type is a milder cut than height or income

Three things keep body type from dominating a result the way height and income can. The measured distribution is broad, so even a fit-leaning preference keeps a large share of the pool if it reaches into the overweight band. The trait is malleable, so the population on any given date is not fixed the way a height distribution is. And stated preferences on build are elastic, since people accept a wider range in practice than they name in a survey, which effectively widens the filter.

Compare that to height, where the distribution clusters tightly around the average and a 6 foot floor alone removes about 85 percent of men, or income, where a six-figure floor removes roughly 82 percent. Those filters are unforgiving because the supply behind them is thin and fixed. Body type is neither. That is why the calculator ranks it among the milder inputs, and why a surprising result almost always traces back to a height or income setting rather than a build one. If your score reads higher than you expected, the guide on how to lower your delusion score shows which filters to loosen first, and body type is rarely the one to blame.

What this means for your result

The practical takeaway is that a body-type preference is reasonable and common, but it interacts with a heavier real population than most people picture. Stating a preference for a fit partner is not delusional on its own. It only starts to strain against supply when it is set strictly, at the athletic tail, and then stacked on top of equally strict height and income demands. At that point the body-type filter stops being the gentle input it usually is and joins the others in shrinking the pool.

You can watch this play out directly. Run a realistic set of filters in the female delusion calculator or the male delusion calculator, then change only the body-type setting and note how little the result moves compared with nudging height or income. The tool models the US adult population using CDC NHANES and Census data, and reports the outcome as a matching percentage, a 1 in X figure, and a score from 1 to 10. For the full source list and the vintage of each dataset behind the body-type estimate, see the methodology page. The broader weight picture that drives this filter is covered in the stats page on obesity rates, which breaks the BMI distribution down by group.