Body Type Calculator

How to measure your body

Four circumferences decide everything on this page, and three of the four are read at a level people routinely get wrong. Use a flexible tape, keep it horizontal all the way round including across the back, and pull it snug without compressing. Measure over bare skin or thin clothing, breathe normally, and do not hold the tape at an angle to make a number come out differently — the whole method compares these figures against each other, so a single misplaced reading distorts the result far more than it looks like it should.

High hip is not the full hip

This is the single most common measurement mistake on any body-shape tool, and it matters here because both measurements are used and they are used differently. The high hip runs across the upper swell of the hip and the upper abdomen, roughly 8–10 cm (3–4 in) below the natural waist, over the top of the hip bones. The full hip is the widest circumference lower down, around the seat. On most bodies the full hip is the larger of the two; on some it is not, and that is exactly the information two of the seven rules are reading. If you take one measurement and enter it in both fields, you are not simplifying the method — you are removing the difference the rules are looking at.

Measure twice

Tape tension, posture and breathing move a reading by a centimetre without any difficulty, and several boundaries in this rule set are narrower than that. Switching on the measure-twice option adds a second field per measurement; where both are filled in, the tool uses their arithmetic mean — (reading 1 + reading 2) ÷ 2 — and prints that averaging step in the receipt, so the number it classified is never a number you cannot trace.

What body shape am I?

The method behind this page is short enough to state completely. Take your four circumferences and compute five quantities from them: bust − hips, hips − bust, bust − waist, hips − waist, and high hip ÷ waist. Then check those five figures against seven independent rules, each of which is a small list of conditions. A rule is satisfied only if every one of its conditions holds; otherwise it is not satisfied, and there is no partial credit.

Three things follow that most body-shape tools quietly skip. First, the seven rules were written independently, so they do not partition every possible body into exactly one category: some measurement sets satisfy two rules and some satisfy none. Second, because of that, the answer is properly a state — one match, several matches, or no match — and each of those three is reported here as what it is. Third, since the rules are checked independently rather than in a fixed order, the sequence they are written in cannot influence what comes out; a tool that tests them one after another and stops at the first hit is reporting an artefact of its own ordering.

The vocabulary and the four-measurement approach come from the Female Figure Identification Technique for apparel, as applied in the cited 2007 comparison of American and Korean women's body shapes. Everything the rules do is compare circumferences; nothing in them observes tissue, health or physiology.

The 7 female body shapes

Every condition below is the actual rule the calculator runs, read from the same versioned dataset. Thresholds are stated in inches because that is the unit the rule set is written in; the centimetre equivalents are generated as inches × 2.54 and are never maintained as a separate rule set.

B = bust, W = waist, HH = high hip, H = full hip. A strict inequality is not satisfied at the boundary itself; a non-strict one is. The tool reports the signed distance from every boundary so you can see which side of each you are on and by how much.

Hourglass vs bottom hourglass vs top hourglass

All three require a waist that is clearly defined against something. What separates them is which of bust and hip runs ahead of the other, and by how much.

Hourglass requires bust and hips to be close to balanced: bust − hips at or below 1.0 in (2.5 cm), and hips − bust below 3.6 in (9.1 cm). On top of that it needs a defined waist, and it accepts either route — bust − waist at or above 9.0 in (22.9 cm), or hips − waist at or above 10.0 in (25.4 cm). That "either" is the only disjunction anywhere in the seven rules, and it is why two bodies can both be Hourglass while getting there through different measurements.

Bottom Hourglass lets the hips run ahead: hips − bust at or above 3.6 in (9.1 cm) but below 10.0 in (25.4 cm), with hips − waist at or above 9.0 in (22.9 cm). It then adds the condition that does the real work — high hip ÷ waist below 1.193 — which is what keeps it distinct from Spoon.

Top Hourglass is the mirror image on the upper body: bust − hips above 1.0 in (2.5 cm) but below 10.0 in (25.4 cm), with bust − waist at or above 9.0 in (22.9 cm). Note that it needs no high-hip measurement at all.

The three are mutually exclusive by construction — the bust − hips condition cuts the same axis three ways — so no body satisfies more than one of them. Overlaps in this rule set happen between an hourglass variant and one of the other four categories, not among the variants themselves.

Pear, spoon and triangle: what's the difference?

"Pear" is not a category in this rule set; it is the everyday name for Triangle, and this page labels the category with both. Triangle and Spoon both describe lower-body-weighted proportions, and they are separated by two conditions rather than by feel.

Triangle needs hips − bust at or above 3.6 in (9.1 cm) and hips − waist below 9.0 in (22.9 cm) — hips clearly ahead of the bust, but without a sharply defined waist. It does not use the high hip at all.

Spoon needs hips − bust above 2.0 in (5.1 cm), hips − waist at or above 7.0 in (17.8 cm), and high hip ÷ waist at or above 1.193. That last condition is the whole point of the category: it says the fullness starts high, immediately below the waist, rather than only at the widest part of the seat.

Because Spoon's waist condition (7.0 in (17.8 cm)) is looser than Triangle's cut-off (9.0 in (22.9 cm)), the two genuinely overlap in the band between them: a body with hips − waist in that range and a full high hip satisfies both rules. This calculator reports both rather than choosing. Worked example: at 36 in bust, 29 in waist, 36 in high hip and 41 in hip, hips − waist is 12.0 in and high hip ÷ waist is 1.241, which satisfies Spoon and fails Triangle. Drop the high hip and the same body becomes Triangle territory instead — which is exactly why the tool refuses to guess that measurement.

By contrast, at 34 in bust, 31 in waist, 33 in high hip and 38.5 in hip the result is Triangle: hips − waist is 7.5 in, under the 9 in cut-off, and high hip ÷ waist is 1.065, under 1.193.

Rectangle vs apple/oval

Rectangle is a category in this rule set. Apple and oval are not, and this is worth being precise about, because the substitution people usually make is wrong.

Rectangle requires four things at once: hips − bust below 3.6 in (9.1 cm), bust − hips below 3.6 in (9.1 cm), bust − waist below 9.0 in (22.9 cm) and hips − waist below 10.0 in (25.4 cm). In words: bust and hips close to each other, and a waist that is not sharply defined against either. It is a statement about all four measurements being relatively close together.

"Apple" and "oval" in everyday use describe fullness concentrated around the midsection — a waist that is large relative to bust and hips rather than merely undefined. That is a different thing from Rectangle, and it is not the same as Inverted Triangle either, which is the substitution this page will not make. Inverted Triangle requires bust − hips at or above 3.6 in (9.1 cm) with bust − waist below 9.0 in (22.9 cm): an upper body clearly larger than the lower one. A body described colloquially as apple-shaped may have bust and hips close together, which puts it nowhere near that condition. Treating apple as a synonym for inverted triangle silently reassigns bodies to a rule they do not satisfy.

What this calculator does instead is state which of the seven rules your measurements actually satisfy, and if none of them does, say so. The FFIT literature includes further categories beyond the seven implemented here; this tool implements seven and does not stretch them to cover names it has no rule for.

Men's body proportion calculator

Switch the mode selector at the top of the page to the general profile and the calculator stops running the seven-shape rules entirely. That is deliberate, and it is the honest answer to "can men use a body type calculator?".

The seven-category rule set was developed and published for female figures in apparel research, with thresholds fitted to female anthropometric survey data. Applying it to a male body would silently reuse thresholds that were never derived for one. This mode therefore reports measured proportions and arithmetic differences only.

What the general mode reports instead is measured: chest ÷ hips, waist ÷ hips, chest − waist drop, hip − waist drop, chest − hips (upper-to-lower balance). Alongside those it prints two descriptive readings — waist definition and upper-to-lower balance — each of which is explicitly a description of the figure directly above it rather than a classification of the body, with its bands stated on screen.

What it does not do is print a shape name. No published, verifiable rule set with explicit numeric thresholds for men's apparel shape labels (trapezoid, oval, and similar) was available at build time. Inventing thresholds and printing a shape name from them would present an unsourced classification as a result, so no shape label is emitted in this mode. The quantitative profile below is reported instead, and every band in it is a description of the visitor's own arithmetic rather than a classification of their body.

If you want a single figure to track over time, the chest − waist drop is the one tailoring actually uses, and it is reported in both units. The arithmetic is fully visible, which is the part that matters: a proportion you can recompute is more useful than a label you cannot check.

Body measurement chart

What each of the four measurements is, where it sits and what the rules do with it. Every threshold that touches a measurement is listed against it.

The thresholds in both units

Every threshold is stored once, in inches (or as a dimensionless ratio). Centimetre equivalents shown anywhere on the page are generated at display time as inches × 2.54 and are never stored, rounded or maintained as a second, independent metric rule set.

Body type, body shape, somatotype and body composition are not the same thing

These four phrases get used as if they were interchangeable, and they are not. Here is what each one refers to and whether four tape measurements can get anywhere near it.

The row that catches people out is somatotype. Ectomorph, mesomorph and endomorph belong to a separate historical framework with different inputs and a different purpose; they are not a finer-grained version of body shape and they are not produced anywhere on this page. If you arrived looking for one of those three words, this calculator will not give you one, because the four measurements it holds cannot support it.

The same applies to body fat, muscle mass and metabolism. A circumference records how far the tape went around; it holds no information about what is underneath it. Any figure claiming to read body fat percentage or metabolic type from bust, waist and hip is asserting more than its inputs allow.

How accurate is a body type calculator?

Accuracy is the wrong frame here, and swapping it for the right one is the most useful thing this section can do. A body-shape calculator is not estimating a hidden quantity that could be measured better — it is applying a published rule set to numbers you supply. The rules either hold or they do not. So the question is not "is the label correct?" but two sharper ones: are the measurements good enough, and how close to a boundary does this body sit?.

Measurement error dominates

Tape tension, posture, breathing and tape level routinely move a circumference by a centimetre. Several boundaries in this rule set are narrower than that, which means measurement error, not the rules, is the main source of a changed answer. That is why the stability panel exists: it reclassifies a fixed grid of nearby measurements and tells you whether a centimetre would have changed the outcome. A result that survives every scenario is a different kind of answer from one that survives nine of sixteen, and you should be told which you have.

The rules have structural limits too

The thresholds are absolute inch distances, so the rule set is not scale-invariant: a 9-inch bust-to-waist difference is a much bigger relative difference on a small frame than on a large one, and two bodies with identical proportions at different sizes can land in different categories. Only one condition in the whole set — the high hip ÷ waist ratio — is dimensionless.

What a tape cannot see

Circumference collapses width and depth into one number, so two bodies with the same measurement can have quite different cross-sections and need different garment patterns. Shoulder geometry, one of the strongest drivers of how a silhouette reads, is not measured at all. Fat distribution changes the readings without the method ever observing the cause.

Compared with what modern anthropometry does

Contemporary body-shape research uses 3D body scanning, working from point clouds with dozens to hundreds of landmark-derived parameters, including the depth and cross-sectional information a tape cannot capture. This page is a four-measurement approximation with a fixed rule set — a deliberate trade of resolution for something you can reproduce at home with a tape measure. It is not a scan, and it should not be read as a low-cost substitute for one.

Where this calculator is genuinely better than a label

Not because its rules are better — they are the same published rules — but because it shows the distance to every boundary, reports overlaps and no-matches instead of hiding them, and never estimates a measurement you did not take. Worked example: at 36 / 26 / 34 / 39 in, two rules hold at once — Hourglass and Spoon — and both are reported. At 44 / 34 / 36 / 34 in, none of the seven holds, and the page says so rather than defaulting to Rectangle.

A description of proportions, not a fact about your body

These categories were built to describe how garments meet external proportions, and that is the whole of what they claim. A shape name says something about the arithmetic relationship between four circumferences on the day you measured them. It says nothing about your body composition, your health, your fitness or your metabolism, and nothing here can be turned into any of those.

Overlap and no-match are real answers

Seven rules written independently over four measurements do not partition every possible body into exactly one box. Some measurement sets satisfy two rules at once, and some satisfy none. A calculator that always returns a single label is hiding one of those two cases, usually by testing the rules in a fixed order and stopping at the first hit. This one evaluates all seven and reports what it actually found.

The boundaries are narrower than your tape

Several of these thresholds are separated by less than the difference between a tape held snugly and a tape held loosely. That is why every result here carries a distance to the nearest boundary and a check across a fixed grid of nearby measurements: if a centimetre would have flipped the answer, you should be told, not reassured.

High hip is a measurement, never an assumption

Two of the seven rules turn on the relationship between the high hip and the waist, and nothing else on the page uses that measurement. If you have not taken it, the tool says which rules it cannot settle and asks for it — because the alternative, estimating a high hip from the measurements you did give, would manufacture the very number the rule depends on.

Methodology and sources

Body Shape Classification Method v1.0 · dataset version 1.0.0 · engine version 1.0.0 · reviewed August 9, 2026.

Where the method comes from

The seven category names, the four measurements they compare and the general form of the conditions follow the Female Figure Identification Technique for apparel, originating with Karla P. Simmons, Cynthia L. Istook, Priya Devarajan (2004), Female Figure Identification Technique (FFIT) for apparel, Part I: Describing female shapes, Journal of Textile and Apparel, Technology and Management 4(1) — and applied in the paper this dataset cites as its primary source:

Jeong Yim Lee, Cynthia L. Istook, Yun Ja Nam, Sun Mi Park (2007). Comparison of body shape between USA and Korean women. International Journal of Clothing Science and Technology, 19(5), 374-391. DOI: 10.1108/09556220710819555.

That citation — title, all four authors, journal, volume, issue, page range, year and DOI — was checked against the publisher's own record on August 9, 2026. Origin of the FFIT shape vocabulary and of the four-girth (bust, waist, high hip, hip) approach this rule set uses. The full text was not retrievable from this environment at verification time; the citation itself is corroborated by multiple independent bibliographic records.

What was and was not verified

The seven category names, the four measurements they compare and the general form of the conditions follow the Female Figure Identification Technique for apparel, as applied in the Lee, Istook, Nam and Park (2007) paper whose DOI is verified above. The specific numeric values (1 in, 2 in, 3.6 in, 7 in, 9 in, 10 in and the high-hip/waist ratio 1.193) were supplied in the CalcDomain implementation brief and are implemented here exactly as given, so that the tool is reproducible against the rule set it claims to use. They were NOT read back from the full text of a primary FFIT publication during implementation: the historical open-access URLs for the 2004 JTATM articles no longer resolve, and the publisher records reachable from this environment expose metadata and abstracts only. This dataset therefore states the thresholds as implemented and does not attribute each individual number to a page of the primary literature. Anyone re-verifying this should read Simmons, Istook and Devarajan (2004) directly and correct this dataset if a value differs — the page, the engine and the tests all read these numbers from here, so a single edit propagates everywhere.

Citation and DOI verified against the publisher record on 2026-08-09 by an automated check. No medical, clinical or expert review of any kind has been performed on this page, and none is claimed.

Limitations

• The thresholds are fixed absolute distances in inches, so they are not scale-invariant: two bodies with identical proportional relationships but different overall sizes can fall into different categories, because a 9-inch bust-to-waist difference is a much larger relative difference on a small frame than on a large one.
• The high hip ÷ waist condition is the only dimensionless rule in the set; every other condition compares absolute inch differences, so the rule set mixes scale-dependent and scale-independent criteria.
• A tape measure records circumference. Two bodies with the same circumference can have very different cross-sectional shapes, because circumference collapses body depth and width into a single number.
• Body depth — how far the body extends front to back at each level — is invisible to this method, and it is one of the main reasons two people classified the same way can need different garment patterns.
• Shoulder geometry is not measured at all, although it is one of the strongest visual drivers of how a silhouette reads.
• Where body fat is distributed changes the measurements without changing anything the method claims to know; the method observes the result, never the cause.
• Posture at the moment of measuring changes the readings: standing tall, breathing in, or letting the tape slope will each move a circumference by more than some of the boundaries the rules use.
• Measurement error is not incidental here. Several boundaries are separated by less than the difference between a tape held snugly and a tape held loosely, which is why every result on this page is accompanied by a sensitivity check.
• The boundaries are sharp: a body sitting a fraction of an inch either side of a threshold receives a different label, even though the two bodies are essentially identical.
• The whole rule set is apparel classification, developed to describe how garments fit external proportions. It is not a medical classification, it is not diagnostic, and no health conclusion follows from any category in it.
• Modern anthropometry uses three-dimensional body scanning and dozens to hundreds of landmark-derived parameters. This is a four-measurement approximation with a fixed rule set, which is a deliberate trade of resolution for something a person can do at home with a tape measure — not an equivalent of a scan.
• The rule set was developed on adult female populations for apparel; it does not carry over to children, and it is not applied to male bodies anywhere in this tool.

Assumptions

• The four inputs are tape circumferences taken on a bare or lightly clothed body, with the tape horizontal and snug but not compressing.
• The seven-shape classifier is applied only in the female mode; the general mode reports measured proportions and never runs those rules.
• All conditions are evaluated in inches, the unit the published rule set is written in. Centimetre entries are converted with the exact factor 1 in = 2.54 cm before anything else happens.
• Every one of the seven rules is evaluated independently against the same measurements, so rule order cannot influence the outcome.
• No measurement is ever assumed. If the high hip has not been entered, the two rules that need it are reported as undetermined rather than resolved with a guessed value.
• Waist-to-hip ratio is computed from the same inputs but is never a condition of any shape rule.

Nothing in this rule set is a medical, diagnostic or body-composition method. A body-shape category describes external proportions for garment fit and carries no information about health, fitness, body fat, muscle, metabolism or risk.