Pear Shaped Face: One Column, Five Bands, and the Square Boundary

One outline, three naming traditions

Start with the geometry, because it does not change with the word used for it. The forehead is the narrowest of the three widths; the cheekbones sit wider below it; the jaw is the widest line on the face, and the chin ends broad or softly rounded rather than coming to a point. A guide that calls this outline a pear is describing its face-on silhouette, narrow at the top and heavy at the bottom. A guide that calls it a triangle is describing the same thing with the base of the triangle resting on the jaw.

A third tradition lists its shapes with the triangle split into two by direction. Base-up triangle is the one with the wide forehead and high cheekbones, narrowing gradually to the chin. Base-down triangle is the one with the narrow forehead that widens through the cheek and chin areas. Base-down is the outline this page is about. Base-up is the outline this site files under heart — the same one other guides call inverted triangle, which is the naming trap running in the opposite direction, and the reason the first thing worth checking in any guide is which width it means when it names a triangle.

On this site all of those names land in existing columns. Seven shapes are listed, pear is not one of them as an eighth category, and no threshold had to be invented to place it: the label points at an outline that already had a column, so this page is a distinction rather than a new profile. The measurements behind the word pear are the triangle column of the table below, and the same is true whether a guide calls it a pear, a triangle, or the base-down half of a pair.

The label is also applied loosely in circulation — some guides use it for any outline that widens downward, others only for the extreme end of that family. This site's rule set does not split the family, so the bands below cover the whole of it; how extreme your own taper measures is a reading the table returns, not a different category.

Three readings of this outline, and the one a tape answers

Guides separate this outline from its neighbours by one of three readings, and the three are not freely interchangeable. One reads the hairline: the line the hair grows along, taken as straight or rounded on one side and centre-pointed on the other. One reads the forehead alone: how wide it measures against the rest of the face. One reads the full gradient: forehead, cheekbones and jaw measured against each other and read as a sequence.

A face can satisfy one of those readings and fail another, and that is not a mistake in any of them — they are different criteria rather than different qualities of the same criterion. A straight hairline over an outline that clearly widens downward is common, and so is a centre-pointed hairline over a face whose widest line is the jaw. Two guides reading different criteria will both report confidently and will disagree, which is exactly what happens across the pages that use this label.

This site reads the third one, and the reason is repeatability rather than preference. A hairline moves: a fringe covers it, recession raises it, and holding the hair back for a measurement moves what a tape can reach. The three widths and the jaw angle are bone-and-tape quantities that do not move between takes, which is what lets them become thresholds — and a criterion that cannot be taken the same way twice cannot become one. That is the whole argument for reading the gradient; it is not a claim that the other two readings are wrong.

If a guide read you by one of the other two and landed on pear, there is nothing to resolve. The label tells you which tradition measured you; the table below tells you what your face measures. Read the triangle column either way.

The table: five checks, two columns

Rows are the five checks this site classifies on. Columns are the triangle column, which is where this label sits, and the square column, the outline guides most often compare it against. The other five shapes are not printed here: the full seven-column version lives on this site's rectangle face shape page, and this page carries two columns because these are the two in play for this label.

Four notes belong to the reading of this table rather than sitting under it as a disclaimer — the boundary section below depends on the first one.

• The bands overlap, and on this pair four rows out of five do. A reading inside a shared zone is placed by the other checks rather than by that row, and at the edges by convention. The boundary section works through which row nearly parts the two columns and which four do not.
• One column is measured and one is not. Square's centre was fitted to a group of six photographs in this site's calibration set; triangle had no photograph behind it at all, so its row sits where the published ordering of the seven shapes puts it. The about page prints the method and both of its limits.
• The jaw angle is the weakest of the five checks in this site's own data: it could not separate round from square — two groups that measured within a degree of each other, though the published literature places the two shapes far apart. Treat the angle as a tie-breaker and the ratios as the evidence.
• Readings from the photo detector arrive on a different scale and are converted before they are shown anywhere on this site. A tape measurement needs no conversion: you are already on the scale this table is printed in.

Taking four of the five numbers at home

Four lengths, one soft tape, bare skin, hair pushed back off the forehead, head level. Face length runs from the midpoint of the hairline straight down the midline to the tip of the chin. Cheekbone width runs apex to apex across the highest points of the cheekbones, the tape passing just under the eyes. Forehead width runs across the widest part of the forehead. Jaw width runs from one jaw corner to the other along the outside of the jaw, straight across rather than wrapping under the chin. No tape at hand? The string and bank-card method described on this site gets the same four marks out of any ruler.

Three divisions turn those four lengths into three of the checks: length over cheekbones, forehead over cheekbones, jaw over cheekbones. The units cancel, which is why centimetres and inches give identical quotients and the table above prints none of either.

The fourth check, the jaw angle, is not one a tape will hand you. On the calculator path this site estimates it from your three ratios and says plainly that it is an estimate; on the photo path it comes from landmarks, in your browser, with nothing uploaded.

The fifth check, chin taper, is one this site deliberately leaves out of the tape path. A chin width is ambiguous by a few millimetres of tape placement, and a few millimetres is most of the band, so the calculator has no field for it and the classifier sets its weight to zero there rather than accepting a number it cannot stand behind. The photo path is the one that produces it, and on this pair that costs you less than it might sound: the boundary with square does not rest on it.

Of the four a tape gives you, the forehead width is the one worth taking twice on this label, with the hair pulled back and the tape at the widest point. It is the check that nearly parts this outline from square, and hair lying over the forehead lowers what the tape reads without changing the face underneath.

The square boundary: one row nearly parts them

More than one guide treats square as the shape to compare this outline against, and this site's own rule set names square as the boundary for it, so the comparison is worth taking apart by row instead of by impression.

Four of the five rows leave the two columns sharing ground — the length bands, the jaw bands, the angle bands and the chin bands each overlap, so a reading sitting inside a shared zone is settled by the other features rather than by that row. The row that nearly parts them is the forehead ratio. The two forehead bands sit edge to edge rather than stacked: take the nearer edge of one away from the nearer edge of the other and the ground the two columns share is 0.01 of the ratio scale — the narrowest shared span anywhere in the five rows. Said in readings rather than numbers: a forehead measuring well under the cheekbone width places you in this column, a forehead measuring level with the cheekbones places you in the square one, and a reading inside that one-hundredth is the single case where this row does not decide.

The jaw row points the same way without separating anything on its own. This is the column whose jaw band sits highest in the table and clears the cheekbone width at its centre, while a square's jaw band sits below the cheekbone width with the three widths running close together — but the two jaw bands still overlap, so the row reads as agreement when the forehead row has already decided, not as a substitute for it.

That is the walkthrough, and it comes with the same limit the table carries: the column this turns on is the one with no photographs behind it, so the forehead row is a definite line in this site's published rule set rather than a line that has been tested against a sample of faces. If the rows disagree with each other on your numbers, the result page reports both shapes with the margin against each rather than picking one for you.

What the readings cannot carry

The five checks are bone measurements. Soft tissue sits on top of the bone — cheek fullness, the jaw muscle, weight change, age — and shifts what a tape reads without shifting the structure under it, which is why two people with the same skeleton can measure differently a decade apart.

Face length is taken from the hairline, so anything that moves the apparent hairline moves that check: recession raises it, a fringe lowers it, and hair worn forward rather than held back changes the top of the measurement without changing the face. The forehead width the boundary above turns on is read the same way, which is why the measuring section asks for the hair to be out of the way on that one.

The jaw angle is an estimate on the calculator path and a landmark reading on the photo path, and in this site's own calibration it was the least dependable of the five — the one place where the two paths can genuinely part company.

The photo path's quotients are produced on a different scale from a tape and are converted before you ever see them, so what a result page shows is an estimate on the tape ruler, not the raw landmark number the classifier consumed. That conversion is a known, published one; it is still a conversion.

None of this is a reason to distrust a reading, and none of it is a caution about anything you do with your face. It is the list of things a band can and cannot carry: what a band can carry is where you sit on a ruler, and what it cannot carry is the soft tissue, the hairline and the angle estimate sitting on top of that ruler.

Weight and lines on an outline that widens downward

This section is a styling judgement and is marked as one: nothing in it comes from the rule set, and none of it is a correction. The full treatment — hair, frames, facial hair, hats, earrings, necklines and makeup placement — is on this site's triangle guide, because it is the same outline under any of the names above, and writing it out twice would only give you two places to disagree with.

The outline carries its weight low: narrowest at the forehead, widest at the jaw. So the styling question is never which line to introduce, it is where weight sits. Weight raised above the cheekbones — volume at the crown, a fringe that fills the forehead, layers beginning at cheekbone height — travels against the gradient. Weight set at or below the jaw — blunt ends landing on the jawline, a frame carrying its mass along the lower rim, a beard squared wide at the corners — adds to the end that already has the most. Both are looks, and the second is a recognisable one with a great deal of good styling attached to it; what the paragraph describes is where the weight lands, not which version is allowed.

The small categories follow the same direction. An earring that ends above the jaw leaves the widest line alone; one that hangs to it marks it. A neckline holding a horizontal at the jaw repeats the widest line; a deep narrow one moves the eye up and inward. A frame with its weight along the top rim puts the heaviest line above the cheekbones, which is the move the whole section keeps making.

The triangle guide's picks are grouped into women's cuts, men's cuts and a beard section, and both groups make the same move on this outline — raise the width, leave the jaw alone. That is also the short answer to whether any of this is written for one audience: the reasoning is about where weight lands, not about who is wearing it.

Frequently asked questions

Is a pear shaped face the same as a triangle face? On this site it is the same column: pear is a name for the outline the rule set files under triangle, and the table above prints that column's bands. The name travels in more than one direction in circulation — some guides say pear, some say triangle, and some split the family into base-up and base-down and take the base-down half for this outline. Nothing about the geometry changes with the word, and the site lists seven shapes rather than adding an eighth.

How do I tell this outline from a square one? Take the forehead width first, with the hair held back. On this outline the forehead is the narrowest of the three widths and its band is the lowest forehead band in the table, shared only with diamond; on a square the forehead runs level with the cheekbones. It is the one row whose two bands barely meet, so a reading at either end of that scale settles the label on its own — and a reading in the shared hundredth does not, which is when the other rows take over.

Is a pear shaped face rare? We do not know, and this site will not guess. No prevalence figure is published here for this or any of the seven shapes, because the figures in circulation carry no stated sample and no stated method to check them against.

Can a face be a pear shape and an inverted triangle at the same time? No — the two run in opposite directions. A pear widens downward to the jaw; an inverted triangle widens upward to the forehead, which is the outline this site files under heart. The base-up and base-down naming says the same thing from the other side of the family, and it is why the first question to ask any guide is which width it means.

Do these recommendations apply to men as well as to women? The reasoning is about where weight sits against an outline, so it is not tied to an audience. The triangle guide's picks are grouped into women's cuts, men's cuts and a beard section, and both groups make the same move on this shape.

Can a pear shaped face change over time? The bone under it does not change in adulthood; the readings can. Weight shifts cheek fullness and jaw definition, age changes soft tissue, and the hairline the length is measured from moves with recession or with a fringe. That is part of why the table above is a band and not a single value.

Why does the detector never return pear? Because seven categories are all it returns, and pear is one of the names this site files inside triangle. The result page prints the margin against every shape, so a triangle result also shows how close round and square sat — which is the useful part of a mixed-shape reading.

I measured with a tape and the photo result shows different numbers. Which is right? Both, each on its own scale. The photo path reads a different set of points, and its quotients are converted to the tape scale before they are displayed; a tape reading needs no conversion, and you are already on the scale the table is printed in. The calculator will not ask you for a chin width either way.

Where the numbers come from

Every band in the table is this site's published rule set: the seven prototypes, their standard deviations, the mapping between the detector's landmark scale and the tape scale, and the weights the classifier runs on. The whole of it is printed on the about page, together with the two limits that matter most here — that the triangle row is interpolated rather than measured, and that the jaw angle could not separate round from square in the calibration data.

The one figure in the boundary section is a single subtraction of two endpoints printed in the table, and the description is written out there so you can redo it. No other number on this page comes from anywhere else, and none of it is estimated: figures that could not be traced back to the rule set have been left out, which is why the page carries no prevalence numbers, no dating of anyone's guidance, and no centimetre targets.

The styling reasoning in the seventh section is ours and is labelled ours. It follows from how the rule set describes this outline rather than from any measured outcome, and it has not been reviewed by a licensed cosmetologist — the about page carries that gap openly instead of inventing a credential to fill it.

What this page does not do: no eighth prototype, no threshold we cannot defend, no rating of your face, and no claim about how this outline photographs. On every path through this site your photo stays on your device.