What's a normal fWHR? Is a +3° canthal tilt average? How symmetric is a typical face? This page collects reference ranges for the measurements our tools report, explains where they come from, and shows how our scanner turns each one into a score.
Where these norms come from, and where they don't
Our numbers come from two places: published research (anthropometry, orthodontics and face-perception studies, listed at the end of this page) and the reference targets our scorer uses, which are set from that literature. They do not come from a database of user scans. We deliberately don't build one: uploaded selfies are deleted within 2 hours and free scan results after 48 hours. When a table below shows a percentile, it's the percentile our scoring model assigns, not a count of real faces.
Norms also vary by age, sex and ancestry, and many studies sample specific populations. Treat every range as a broad reference, not a standard you should meet.
Proportions
| Measurement | Definition | Typical / reference range | Scanner target |
|---|---|---|---|
| Length-to-width | upper forehead–chin ÷ cheekbone width | ~1.3–1.6 depending on the forehead landmark | 1.45 ± 0.14 |
| Facial thirds | three vertical sections ÷ length | rarely equal; a few percent difference is normal | 33% each |
| Facial fifths | five horizontal segments ÷ width | each roughly one eye width | 20% each |
| Midface ratio | pupil distance ÷ pupil-to-lip height | ~0.9–1.1 | 1.00 ± 0.08 |
| Eye spacing | pupil distance ÷ cheekbone width | ~0.43–0.49 | 0.46 ± 0.03 |
| Nose width | nose width ÷ inner-eye gap | ~0.85–1.15 | 1.0 ± 0.16 |
| Mouth-to-nose | mouth width ÷ nose width | ~1.3–1.8 | 1.55 ± 0.22 |
Classical canons say the thirds and fifths should be equal. Anthropometric measurement, starting with Farkas, shows few real faces match them exactly, which is why our scorer uses tolerance bands rather than exact targets.
Facial width-to-height ratio (fWHR)
fWHR divides cheekbone width by upper-face height. Published values depend heavily on the height definition (brow to upper lip vs. nasion to upper lip vs. eyelid to lip), so compare numbers only within one method. With our definition (brow center to upper lip), typical values fall around 1.8 to 2.0, and our target is 1.9 ± 0.18. Research such as Carré and McCormick (2008) found higher fWHR associated with perceived and measured aggression in specific samples; later work has debated how general that link is.
Canthal tilt
Studies of eyelid morphology generally report a mildly positive average intercanthal inclination of a few degrees, with differences between populations and between sexes. Left and right eyes commonly differ by 1–3°. Our scorer centers on +5.5° with a 4° tolerance, so tilts from about +2° to +9° score well, and the canthal tilt calculator reports each eye separately after levelling the head.
Symmetry
There's no universal symmetry scale; studies use different landmark sets and normalizations. On ours (12 bilateral pairs, normalized by half the cheekbone width, score = 100 × e^(−3.2 × D)), clear frontal photos usually score 80–93. Research finds a modest link between symmetry and attractiveness ratings, and that everyday asymmetry is rarely noticed at a glance. Photo conditions matter: Ward et al. (2018) showed that close-range selfies measurably distort facial proportions.
Jaw
| Measurement | Typical / reference range | Scanner target |
|---|---|---|
| Jaw-to-cheek | ~0.70–0.90 | 0.80 ± 0.09 |
| Gonial angle (lateral X-ray) | roughly 120–130° in adults | n/a |
| Jaw angle (front-photo projection) | own scale; lower = sharper | 125° ± 14° |
The true gonial angle is measured in profile; our front-photo jaw angle is a projection and runs on its own scale.
Profile angles
| Measurement | Reference | Source |
|---|---|---|
| Facial convexity (180° − G–Sn–Pg′) | about 12° ± 4° | Legan & Burstone (1980) |
| Nasolabial angle | about 90–110° | widely cited clinical range |
| Upper lip to E-line | about −4 mm | Ricketts (1968) |
| Lower lip to E-line | about −2 mm | Ricketts (1968) |
Profile norms vary considerably by ancestry; fuller lips forward of the E-line are typical in many populations. Our side profile analysis estimates these from the 3D mesh, using the average adult iris diameter (≈ 11.7 mm, Rüfer et al. 2005) as a millimetre scale.
How to measure yourself consistently
Most variation between two scans of the same face comes from the photo, not the face. To track a measurement over time:
- Use the same camera, distance (arm's length or more) and height (eye level) every time.
- Use the same light, ideally facing a window at the same time of day.
- Keep a neutral expression with your mouth closed; smiling changes canthal tilt, jaw ratios and symmetry.
- Pull your hair back so the forehead and cheek edges are visible.
- Take three photos and compare the median rather than a single result.
How the scanner turns a measurement into a score
Each measurement is scored with a Gaussian closeness curve: score = 100 × e^(−½ × ((value − target) ÷ tolerance)²). A value on target scores 100, one tolerance away about 61, two tolerances away about 14. The 11 scores are combined with fixed weights (symmetry 20%, thirds 14%, canthal tilt, length-to-width and midface 10% each, and smaller weights for the rest) into a harmony score from 35 to 98. Modelled percentiles map each score through a normal curve with mean 60 and SD 18.
Measure yourself
Run the free golden ratio calculator for proportions, the jawline rating test for fWHR and jaw, the face symmetry test and the side profile analysis. Everything runs in your browser.
Free tool
Measure it on your own face
Compare your face with the classical proportion system: golden-ratio checks, vertical thirds, horizontal fifths, fWHR and midface ratio.
Open Golden Ratio CalculatorFrequently asked questions
What is the average fWHR?
It depends on how face height is defined. With brow-to-upper-lip height, typical values are around 1.8 to 2.0.
What is the average canthal tilt?
Studies generally report a mildly positive average of a few degrees, varying by population and sex. Neutral and slightly negative tilts are also common.
How symmetric is the average face?
On our 0–100 scale, clear frontal photos usually score 80–93. No real face is perfectly symmetric.
Are these norms based on your users' scans?
No. They come from published research and the reference targets our scorer uses. We don't keep a database of user faces.
Do facial norms differ by ethnicity?
Yes. Many reference values, especially profile norms, were developed in specific populations. Use ranges as broad references, not standards.
What is a normal gonial angle?
On lateral X-rays, adult gonial angles average roughly 120–130°. Front-photo jaw angles are projections on their own scale.
References
- Farkas LG (1994). Anthropometry of the Head and Face (2nd ed.). Raven Press, New York.
- Carré JM, McCormick CM (2008). In your face: facial metrics predict aggressive behaviour in the laboratory and in varsity and professional hockey players. Proceedings of the Royal Society B 275:2651–2656.
- Rhee SC, Woo KS, Kwon B (2012). Biometric study of eyelid shape and dimensions of different races with references to beauty. Aesthetic Plastic Surgery 36(5):1236–1245.
- Thornhill R, Gangestad SW (1999). Facial attractiveness. Trends in Cognitive Sciences 3(12):452–460.
- Ward B, Ward M, Fried O, Paskhover B (2018). Nasal distortion in short-distance photographs: the selfie effect. JAMA Facial Plastic Surgery 20(4):333–335.
- Ricketts RM (1968). Esthetics, environment, and the law of lip relation. American Journal of Orthodontics 54(4):272–289.
- Legan HL, Burstone CJ (1980). Soft tissue cephalometric analysis for orthognathic surgery. Journal of Oral Surgery 38(10):744–751.
- Rüfer F, Schröder A, Erb C (2005). White-to-white corneal diameter: normal values in healthy humans obtained with the Orbscan II topography system. Cornea 24(3):259–261.
- Holland E (2008). Marquardt's Phi mask: pitfalls of relying on fashion models and the golden ratio to describe a beautiful face. Aesthetic Plastic Surgery 32(2):200–208.