Glossary

This glossary defines every facial measurement and looksmaxxing term used anywhere on this site, each in one place, each with a single definitive sentence and a link to the page that measures or discusses it.

Every term below traces back to the same 55-entity model this site uses for its own scoring and schema markup, so a definition here never contradicts what a tool page or guide says elsewhere. Where a page measures the term directly, the first mention links to it.

Score, Scale, and Interpretation

Facial Attractiveness

Facial attractiveness is a measurable construct built from seven components: symmetry, facial proportions, facial harmony, averageness, sexual dimorphism, apparent age and skin quality.

This site's composite scan that reads all of them at once scores five of those seven from one photo and names the remaining two as not yet scored, each with its reason stated beside it.

Attractiveness Score

An attractiveness score is the single 1 to 10 number this site's scan returns for one photo.

It is a summary value built from the components above, not a separate measurement of its own, and the test that produces it lists every component behind the number.

Attractiveness Scale (1 to 10)

The attractiveness scale is the 1 to 10 range every score on this site is reported against, divided into named bands rather than a single pass or fail line.

Every band on the full scale is explained on its own page, with the reasoning behind each one, not repeated here.

Percentile

A percentile states where one score ranks inside a stated reference population, not the score itself.

This site withholds it for now, until a real reference population and a test-retest study both exist to support one, a limitation stated plainly on the accuracy and limits page.

Confidence Band

A confidence band is the range around a reported score that reflects measurement uncertainty, not a second score.

Like the percentile above, this site withholds it for now rather than publish a number it cannot yet stand behind, and why this site withholds both for now is stated directly, not buried.

The Component Metrics

Facial Symmetry

Facial symmetry measures how closely the left and right sides of a face mirror each other around the midline, using paired landmark points rather than a visual impression.

Research finds it a reliable but modest predictor of rated attractiveness, and this site's own symmetry test reports it as a percentage, not a pass or fail line.

Facial Proportions

Facial proportions are the size relationships between regions of a face, most often stated as facial thirds, facial fifths, or a width-to-height ratio.

This calculator measures several of them at once, each against the specific reference that proportion is actually published against.

Facial Harmony

Facial harmony is how closely a set of already-measured features agree with each other, not the quality of any single feature scored alone.

It is reported as agreement, the spread between components, and level, their average, on this site's own harmony test.

Averageness

Averageness is how closely a face's proportions sit to the mathematical average of a population of faces.

Research finds composite faces built by averaging many individual faces together rated as more attractive than almost every face averaged into them, which does not make average an insult or a target.

This site's scan does not yet score averageness, because no stated reference population exists yet to average against, a limitation explained alongside the rest of the scale.

Sexual Dimorphism

Sexual dimorphism is how strongly a face carries features typed as male or female.

Research finds it a reliable but modest predictor of rated attractiveness, and one study found female raters preferring slightly feminized male faces over masculinized versions of the same face.

This site's scan does not yet score dimorphism either, for the same reason it withholds averageness, and it sits in that same scale as a named, unscored component.

Apparent Face Age

Apparent face age is how old a face reads from one photo, based on skin texture, contrast and facial volume, not a fact about a birth date.

This site's own age guesser returns one estimate per photo, best read as a single data point rather than a verdict.

Skin Quality

Skin quality covers texture, tone and complexion as they read in one photograph, scored on a non-clinical scale rather than a dermatological diagnosis.

No dedicated skin-quality tool has shipped on this site yet; the component is named here because it is one of the seven the composite score is built from.

Proportion and Ratio Terms

Golden Ratio (Phi)

The golden ratio, written phi and equal to about 1.618, is a mathematical ratio some facial-analysis conventions compare a face's height and width against.

Multiple reviews find no reliable link between conformity to phi and rated attractiveness, so this calculator measures the ratio as a descriptive convention, not a score.

fWHR (Facial Width-to-Height Ratio)

fWHR divides a face's bizygomatic width, the width across the cheekbones, by its upper-face height, measured from the brow line to the top of the upper lip.

A meta-analysis found a small but significant link between men's fWHR and aggressive behavior, and the same calculator that reads phi reports fWHR alongside it.

Facial Thirds

Facial thirds divide a face into three horizontal bands, hairline to brow, brow to nose base, and nose base to chin, each measured as a share of total face height.

Published match rates for equal thirds run low in every measured population, and the thirds test on this calculator reports the actual percentages rather than a pass or fail line.

Facial Fifths

Facial fifths divide a face into five vertical columns, each about one eye's width, so the gap between the inner eye corners should equal one eye's own width under the convention.

This calculator's own fifths reading states how closely a face actually matches that convention, which most measured faces do not.

Midface Ratio

Midface ratio divides midface length, from the middle of the eyes to the corner of the mouth, by midface width.

It is a descriptive proportion with no validated aesthetic target, reported for reference alongside the golden ratio and fWHR readings on the same scan.

Midsagittal Line

The midsagittal line is the vertical midline drawn through the center of a face, the axis every symmetry measurement compares the left and right sides against.

This site's symmetry test fits that line to a photo before it measures anything else.

Eye Region Terms

Canthal Tilt

Canthal tilt is the angle of the line from the inner corner of the eye to the outer corner, read against horizontal.

Positive tilt means the outer corner sits higher, negative means it sits lower, and this site's own test reports the actual angle in degrees rather than a label alone.

Positive Canthal Tilt

A positive canthal tilt means the outer corner of the eye sits higher than the inner corner, reported as a positive number of degrees.

Upturned eyes is the everyday name for it, and most published adult averages sit above zero.

Negative Canthal Tilt

A negative canthal tilt means the outer corner of the eye sits lower than the inner corner, reported as a negative number of degrees.

Downturned eyes is the everyday name for it, and the sign describes direction only, not severity.

Neutral Canthal Tilt

A neutral canthal tilt means the two eye corners sit level, or close enough that a 2D photo estimate cannot separate them from a slightly positive reading.

Camera height alone can move a face across that boundary, which is one reason a single photo estimate is not the clinical measurement.

Eye Shape

Eye shape describes the visible outline of the eye, categories such as almond, hooded, upturned, downturned and deep-set.

It is a separate descriptor from canthal tilt, which measures one angle only, and no published typology of eye shape carries a validated population distribution.

Hunter Eyes

Hunter eyes is forum shorthand for a deep-set eye under a low, hooded upper lid, and it bundles four separate traits into one label.

Canthal tilt is one of those four; low eyelid exposure, deep-set orbits and a close brow are the other three, and only the tilt among them is something this site measures in degrees.

Jawline and Cheek Contour Terms

Jawline

The jawline is the lower-face contour running from the chin to the ear along the mandible.

Its perceived definition is read from the gonial angle and jaw width together, and this site's own jawline test reads both from one photo and reports a published Sharp, Average or Soft band.

Gonial Angle

The gonial angle is the angle where the body of the jaw meets its ascending ramus, the bone rising toward the ear, measured at the corner point called the gonion.

A narrower angle near 120 degrees reads as sharper; an angle at 130 degrees or more reads as softer, and this test reports the actual degree value rather than the label alone.

Ogee Curve

An ogee curve is the S-shaped contour running from the cheekbone into the cheek hollow and jaw, a convex-to-concave line borrowed from architecture to describe a smooth midface transition.

No validated aesthetic target for this curve has been published, and no tool on this site scores it directly.

Chewing-Gum Jawline Claim

The chewing-gum jawline claim holds that chewing gum sharpens the jaw.

What the mechanism actually supports is narrower than the claim: sustained chewing can hypertrophy the masseter muscle, which widens the lower face rather than defining it, and a wider muscle is not the same outcome as the angle this claim gets measured against.

Face Shape Classes

Face Shape

Face shape is a descriptive classification based on four measurements: forehead width, cheekbone width, jaw width and hairline-to-chin length.

This site's own detector compares those four points directly rather than relying on a visual guess, which studies find inconsistent even between repeated looks by the same viewer.

Oval Face Shape

An oval face shape has a length visibly greater than its width, a jaw narrower than the cheekbones, and a gently rounded hairline.

Round Face Shape

A round face shape has width and length close to equal, soft angles at the jaw, and full cheeks as its widest point.

Square Face Shape

A square face shape has a strong, angular jawline close in width to the forehead and cheekbones, with minimal taper from top to bottom.

Heart Face Shape

A heart face shape has a wide forehead and cheekbones tapering to a narrow, often pointed chin.

Diamond Face Shape

A diamond face shape has the cheekbones as its widest point, narrowing at both the forehead and the jaw.

Oblong Face Shape

An oblong face shape is longer than an oval, with roughly parallel sides at the forehead, cheek and jaw and minimal curve.

Triangle Face Shape

A triangle face shape has a jaw wider than the forehead, narrowing as it rises, sometimes called a pear shape.

Side Profile Terms

Side Profile

A side profile is the lateral view of a face from forehead to chin.

It shows the front-to-back projection of the forehead, nose, lips and chin that a straight-on photo cannot capture, and this site's own profile test classifies that projection into three named types.

Facial Convexity

Facial convexity is the angle between the forehead, the base of the nose and the chin, and it sorts profiles into convex, straight or concave classes.

A convex profile has a chin set behind that line, a concave profile has a chin set ahead of it, a straight profile sits close to it, and this site's test reports the actual angle behind each label.

Measurement Method Terms

AI Face Analysis

AI face analysis is a method that maps facial landmark points from a photo and measures geometry directly from their coordinates, rather than from a rater's visual impression.

Commercial facial-analysis systems have shown error rates as high as 34.7% for darker-skinned women against far lower rates for lighter-skinned subjects, and this site discloses its own reference population for that exact reason.

Landmark Mesh

The landmark mesh is the set of 478 points this site's scan maps across a face from one photo, 468 for the face itself plus 10 for iris refinement.

Every measurement on this site, from symmetry to canthal tilt to jawline, is read from the same mesh built by the flagship scan, not a separate scan per tool.

On-Device Processing

On-device processing means a photo is measured inside the visitor's own browser and is never uploaded to a server for the free scan.

This site's privacy page states exactly what that claim covers, including that a paid report photo is deleted within 24 hours and never used to train any model.

Other Feature Descriptors

Lip Shape

Lip shape describes the visible outline of the lips, categories such as full, thin, heart-shaped, wide and bow-shaped, read from one photo rather than measured against a numeric standard.

Nose Shape

Nose shape describes the visible outline and profile of the nose, categories such as straight, button, Roman, aquiline and snub, again a descriptive class rather than a numeric measurement.

Looksmaxxing and Rating Vocabulary

Looksmaxxing

Looksmaxxing is the practice of systematically working through a set of appearance changes to raise how attractive a face reads, ranging from sleep and grooming at one end to surgery at the other.

This site's own evidence review grades each named practice by what the research actually supports, not by what the practice claims for itself.

Softmaxxing

Softmaxxing is the reversible, non-medical half of looksmaxxing: sleep, skin care, grooming, body composition, posture and styling.

It is also the half with the strongest published evidence behind it.

Hardmaxxing

Hardmaxxing is the permanent, clinician-delivered half of looksmaxxing: orthodontics, jaw surgery, rhinoplasty, fillers and hair transplants.

Its effects are real, its recovery is real, and none of it can be undone.

PSL Scale

The PSL scale is a 1 to 8 face-rating ladder that formed on looksmax forums, with 4 as its own disclosed midpoint and named after the three forums where it started.

It records one self-selected community's taste, not a population measurement, and it shares its initials with the Pakistan Super League cricket competition by coincidence only.

The full band-by-band breakdown uses numbers and plain descriptions only, not the community's own tier names.

Mewing

Mewing is holding the tongue against the roof of the mouth in an attempt to reshape the jaw over time.

No high-level evidence supports a skeletal effect in adults, and Wikidata's own classification describes it as a form of do-it-yourself oral posture training associated with online looksmaxxing communities, a status this site's evidence review states rather than softens.

Terms Named Here Only, With a Warning

Three terms in looksmaxxing discussion describe documented harm rather than a measurable practice, and this site names them for exactly that reason: so a search for the word returns an accurate definition next to the risk, instead of silence or, worse, instructions. None of the three appears as a heading, an anchor or a how-to anywhere else on this site.

Bonesmashing

Bonesmashing is the practice of striking facial bone on purpose in an attempt to provoke bone growth.

Warning: the practice has no scientific basis, and documented outcomes include fracture, nerve damage and permanent disfigurement; no method for it appears anywhere on this site.

Blackpill

Blackpill is the fatalistic belief that appearance fixes a person's worth and life outcome without recourse.

Warning: researchers studying a major looksmaxxing forum documented dozens of instances of users encouraging self-harm and suicide tied to this belief, which this site treats as a documented mental-health risk, not a competing viewpoint.

Mogging

Mogging is forum slang for one person's appearance visibly outranking another's in a direct comparison, usually posted as a photo-versus-photo callout.

Ranking strangers this way has no measurement behind it and sits in the same harm-flagged category as blackpill, which is why it is defined here rather than modeled anywhere else on this site.

A definition tells you what a term means. It does not tell you where your own face measures against any of them, which is what the free scan on this site is for.

What we won't show you: a confidence band (not until our test-retest study ships) · a percentile (not until a reference population is published) · a skin score from one photo (lighting lies). Every "not scored" comes with its reason.

Questions people actually ask

What is this glossary for?

This glossary defines every facial-measurement and looksmaxxing term used anywhere on this site in one place, each with a single definitive sentence and a link to the page that measures or discusses it.

Is canthal tilt the same thing as eye shape?

No: canthal tilt measures one angle in degrees, while eye shape is a broader visual category such as almond, hooded or deep-set that has no numeric measurement behind it.

Does a PSL score mean the same thing as this site's 1 to 10 attractiveness score?

No: the PSL scale is a 1 to 8 ladder built by one self-selected forum community, while this site's 1 to 10 score comes from a disclosed, published measurement formula, and the two numbers do not convert into each other.

What is the difference between softmaxxing and hardmaxxing?

Softmaxxing covers reversible, non-medical changes such as sleep, grooming and skin care, while hardmaxxing covers permanent, clinician-delivered changes such as surgery, and whether a change can be undone is what separates the two lanes.

Is mewing supported by evidence?

No high-level evidence supports mewing's claimed skeletal effect in adults, and this glossary defines it as a documented forum practice, not as instruction.

Why does this glossary define blackpill and bonesmashing if the site does not endorse them?

Both terms describe documented harm rather than a measurable practice, and this site names them here, with a warning, so a search for the word returns an accurate definition instead of silence or instructions.

Where do I get an actual score instead of a definition?

The free attractiveness test measures your own face against many of these same terms in under a minute, on your device, at no cost.

Sources

Every claim above, traceable

  1. Rhodes G., Annual Review of Psychology 2006 - symmetry, averageness and sexual dimorphism are reliable but modest predictors of rated attractiveness.
  2. Langlois J.H. and Roggman L.A., Psychological Science 1990 - digital composites of 16 and 32 faces were rated more attractive than almost all the individual faces averaged into them.
  3. Perrett D.I. et al., Nature 1998 - female raters preferred slightly feminized male faces over masculinized versions of the same faces.
  4. Naini F.B., Maxillofacial Plastic and Reconstructive Surgery 2024 - review describing the golden ratio's claimed link to facial esthetics as unsupported by the weight of the evidence.
  5. Haselhuhn M.P., Ormiston M.E. and Wong E.M., PLOS ONE 2015 - meta-analysis found a small but significant association between men's fWHR and aggression.
  6. Bashour M. and Geist C., Ophthalmic Plastic and Reconstructive Surgery 2007 - female faces with accentuated medial canthal tilt were preferred 93% of the time over unmodified faces (p < 0.0001).
  7. Farkas L.G. et al., Plastic and Reconstructive Surgery 1985 - measured neoclassical facial canon match rates in young adult North American Caucasians, including 40% and 37% for two horizontal canons.
  8. Halpin C., Gosse M., Yeo E., Handlovsky I. and Maguire J., Sociology of Health and Illness 2025 - a looksmaxxing forum drawing 6 million monthly visitors, with dozens of documented instances of users encouraging self-harm and suicide.
  9. Buolamwini J. and Gebru T., PMLR 2018 - commercial facial-analysis systems showed error rates up to 34.7% for darker-skinned women against 0.8% for lighter-skinned men.

It measures geometry, not worth. The number tells you where you start. It never tells you what you are.

The standing rule
on every page
of this site