The jawline is the feature with the most at stake. Across 17,032 AI face analyses run between January and August 2026, people scoring 8.0 or above overall average 8.89 on the jawline, while people scoring under 5.0 average 4.53. That is a gap of 4.36 points, the widest separation of any of the twelve measured features.
It is also one of only a handful of features marked improvable "Partly" rather than "No". Both of those facts are true at once, and holding them together is the whole point of this guide: the jawline is where the biggest difference between high and low scorers sits, and only part of that difference is available to you.
The jawline is the widest-spread feature in the data
Every analysis scores twelve features separately. Ranking them by the distance between what 8+ scorers average and what sub-5 scorers average puts the jawline first.
| Feature | 8+ average | Under-5 average | Gap | Improvable |
|---|---|---|---|---|
| Jawline | 8.89 | 4.53 | 4.36 | Partly |
| Bone structure | 8.90 | 4.58 | 4.32 | No |
| Skin | 8.33 | 4.05 | 4.28 | Yes |
| Facial harmony | 8.66 | 4.55 | 4.11 | Barely |
| Chin | 8.48 | 4.43 | 4.05 | No |
| Cheeks | 8.25 | 4.50 | 3.75 | Partly |
| Nose | 8.01 | 4.51 | 3.50 | No |
| Forehead | 7.66 | 4.84 | 2.82 | Partly |
The jawline is also the most variable feature outright. Its standard deviation is 1.24, the highest of the twelve, against 0.78 for the forehead and 0.93 for the nose. In plain terms: jaw scores are spread further apart than any other feature, so knowing that the average is 5.59 tells you unusually little about what any individual will score.
That variance is why the jawline dominates looksmaxxing discussion. It genuinely is the feature that separates people most. The mistake is assuming that because it separates most, it is also the feature most available to change.
What a jawline is worth when you have one
A good jaw is a real asset in this dataset, unlike some features that only ever cost points. It is someone's single best-scoring feature 16.0% of the time, which is second only to the eyes at 20.1%, and it is their worst feature 9.1% of the time. Both directions are live.
17.4% of people score 7 or above on the jawline specifically. That is the third-highest of the twelve, behind eyes at 22.6% and eyebrows at 20.2%, and well ahead of the nose at 7.7%. So a strong jaw is not rare in the way a strong overall score is rare.
One number is worth reading carefully. The jawline correlates with the overall score at 0.92, which sounds high but is among the lower figures in the table - facial harmony sits at 0.99, the nose at 0.97, bone structure and the chin at 0.96. A lower correlation means the jaw score moves somewhat independently of the overall number. Practically, that is the difference between a feature that reflects your face and a feature that defines it. The jaw is closer to the former than its reputation suggests.
What actually changes how a jaw scores
Three things move, and they are all soft tissue or measurement rather than bone.
The first is body composition. Facial adiposity - how much fat the face carries - measurably changes how faces are judged; Coetzee, Perrett and Stephen (2009) found perceived facial weight predicted both attractiveness and health ratings. If you carry excess body fat, general fat loss tends to reveal definition that was already there. The honest limit is that this lever is spent once you are lean. A leaner face does not have a different mandible.
The second is short-term fluid and puffiness. Sleep, salt, alcohol and time of day change how sharp the jaw and the area beneath it look on a given morning. This is less a cosmetic trick than a warning about measurement: two photographs of the same face a day apart can genuinely score differently on this feature, and given that the jaw is the highest-variance feature in the data, it is the one most likely to bounce.
The third is the photograph. Camera height, head tilt, focal length and lighting direction all change the visible jaw outline more than they change most features. A camera below eye level softens the jaw line; overhead light flattens the shadow that defines it. Fixing this does not improve your jaw, it improves the measurement of it, and that distinction matters if you are retesting to track progress.
What does not work
Chewing devices and hard gum build the masseter, a muscle. That is a soft-tissue change to how wide the jaw region reads, not a structural one, and it is not a route to a defined jaw for someone carrying facial fat. Mewing, meaning resting tongue posture, is widely presented online as an established method for reshaping the adult jaw. It is not. There is no credible evidence that tongue posture remodels adult facial bone, and the confident claims made for it are marketing rather than findings. Facial exercises do not change a skeleton, and "bone smashing" is self-injury.
This guide does not recommend surgery, and a face-scoring number is not a reason to consider one. If that is a genuine question for you, it belongs with a board-certified surgeon looking at your anatomy, not with an app.
The realistic read
The jawline is the feature with the widest spread and one of the few marked partly improvable, and those two facts pull in opposite directions. The 4.36-point gap between 8+ scorers and sub-5 scorers is mostly structural. The slice you can act on is body composition, sleep, posture and photography: real, worth doing, and measured in fractions of a point rather than whole numbers.
To see where your own jaw sits against the 5.59 average and the 17.4% who clear a 7, run the jawline rating tool, which scores this region on its own. If you want the jaw score in the context of the other eleven features, which is the only way to tell whether it is actually your weak point, the full face rating covers all twelve. And if the jaw turns out not to be your lowest feature, the sibling chin rating covers the region directly beneath it, which is scored separately and behaves very differently.