Jaw width ratio: jaw versus cheekbone width

What jaw width ratio is

Jaw width ratio compares the width of your jaw at the corners, where the jaw turns up toward the ear, with the width of your cheekbones. A value close to 1.0 means a jaw nearly as wide as the cheeks, which reads strong and square. A lower value means a jaw that tapers inside the cheekbones, which reads lean and refined.

In styling convention a wider jaw leans masculine and a tapered jaw leans feminine, so LookFlex uses a different range for each. Both are ranges inside the natural spread of human faces, and the numbers describe proportion rather than strength or beauty.

In the LookFlex app this measurement appears as Jaw width.

How LookFlex measures it

LookFlex measures the distance between the two jaw-angle points, one on each side, and divides it by the distance between the two widest cheek contour points at eye level. Both are horizontal widths, so the measure is mostly sensitive to head turn, which LookFlex corrects for when the 3D fit is reliable.

Because the jaw points sit on the skin contour, soft tissue below the jaw can blur them. The measure therefore starts with high confidence and drops when the head is turned, the light is poor or a heavy beard hides the jaw edge.

The calculation: ratio = width between jaw-angle points ÷ face width at eye level.

Landmarks used

LookFlex reads these points from the 478-landmark face mesh. The numbers are each point's index in that mesh.

• Jaw-angle points: 58, 288
• Widest points of the face at eye level: 234, 454

The full pipeline, including how pose and mirror-averaging reduce noise, is on how LookFlex measures your face.

Typical and scoring ranges

LookFlex shows two ranges. The typical range is the middle half of its reference set, from the 25th to the 75th percentile: half of the faces fell inside it. The scoring range is where a reading earns a closeness score of 100; outside it the score falls smoothly, with no cliff at the edge. Both differ for men and women, and LookFlex blends the two if you choose Other or prefer not to say. Unit: unitless ratio.

How to read your result

"The range" here means the scoring range. The impressions below are common styling descriptions, not research findings. What research does show is in the next section.

• Below the range: The jaw tapers well inside the cheekbones: a lean lower face that, in women, leans toward a heart shape and, in men, toward a V shape.
• Inside the range: Jaw width sits inside the scoring range for your sex.
• Above the range: The jaw is nearly as wide as the cheeks: a strong, squarer frame.

What the research says

Limited evidence Jaw width tracks body mass, and preferences for jaw and face shape vary between people.

• Cheek-to-jaw width and outline measures tracked body mass index (Coetzee et al., 2010).
• Preferences for masculine and feminine facial features are well documented but vary between people and with context (Little et al., 2011).
• Slightly feminised versions of both male and female faces were often preferred (Perrett et al., 1998), and masculine preferences were much weaker in less urbanised societies (Scott et al., 2014).

No study we cite tests LookFlex's own ranges. They come from the calibration described on the methodology page, and every source is listed under References below.

What influences it

Mostly fixed

• The width and flare of the jawbone at its angles, set largely by genetics.
• Muscle bulk at the jaw angle. The chewing muscle varies naturally between people.

Things that move the reading

• Body composition. Across people, cheek-to-jaw width and outline measures track body mass index (Coetzee et al., 2010), so fat along and below the jaw can move the measured width.
• Beard and stubble, which change where the visible edge of the lower face sits.
• Haircut. Side volume and the length at the neck change how wide the jaw looks.
• Posture, which changes the angle between the jaw and the neck in photos.

How to improve how it reads

These suggestions change how the measurement reads or how well the photo captures it. Styling tips are common practice rather than tested in research, and none of them promises a change in bone structure.

• Keep a steady routine of sleep, enough protein and resistance training; the WHO advises muscle-strengthening on two or more days a week (Bull et al., 2020). Gradual change in body composition is the safe lever to try.
• Keep a tidy neckline if you wear a beard, with a clear line under the jaw to show its angle.
• Choose side lengths that fit your goal. Tighter sides show jaw width, fuller sides soften it.
• Skip jaw trainers and hard, constant chewing. We found no evidence that they change jaw width, so LookFlex does not suggest them.

Common myths

Myth: A wider jaw is always more attractive.

Conventions lean different ways for different faces, which is why the ranges are 0.885–0.940 for men and 0.865–0.905 for women. Preferences for masculine or feminine facial features vary between people (Little et al., 2011) and were much weaker in less urbanised societies (Scott et al., 2014).

Myth: Jaw exercises or gum can widen your jaw.

We found no evidence that chewing exercises or gum change the jawbone, and LookFlex does not recommend them as a way to reshape your face.

Myth: Surgery or extreme methods are the only way to change it.

We found no evidence that any home method changes the jawbone, and LookFlex never recommends practices such as 'bone smashing'. Haircut, grooming and body composition change how the jaw reads.

How it feeds your score

This measurement is one input to one sub-score, which then combine into your Glow Score:

• Jawline: 20% of the sub-score. Jawline carries 17% of the Glow Score.

Precision. Every reading of this measurement comes with a 95% margin, estimated by measuring your photo three times. Across the reference set the median margin was ±0.0071: a retake in similar conditions usually lands within that much of the first reading. A change between two scans counts as real only when it is bigger than both scans' margins combined.

Confidence. LookFlex starts this measurement at a confidence of 85% and lowers it when your head angle, the lighting or image quality weakens the reading. The weight of each input in a sub-score scales with its confidence. Details are on the methodology page.