What actually gets measured
"Handsome" is not one thing, which is why a single number on its own is close to useless. The reading breaks your face into 17 separate proportions and shows each one individually, so you can see what is driving the result instead of guessing.
Four of them carry more weight in how male faces are read, according to the published literature:
Jaw & gonial angle
The angle where the jaw turns up toward the ear. Drives how defined the lower face reads. Body composition and camera angle both move it more than people expect.
Facial width-to-height ratio
Cheekbone width against upper-face height. The Grammer & Thornhill 1994 dominance signal — one of the most studied male-face measurements there is.
Brow position & canthal tilt
Where the brow sits relative to the eye, and the angle between inner and outer eye corners. Together these set how the upper face is read.
Chin projection & facial thirds
How far the chin comes forward, and whether the three horizontal thirds of the face are balanced. Structural, but photo angle distorts both heavily.
Most low scores are the camera, not the face
This is the single most common finding and it is worth stating before you scan. A phone at arm's length uses a wide-angle lens. It enlarges the nose, flattens the midface, and shortens the jaw. The measurements read those distortions as real proportion problems, because from the photo's point of view they are.
Overhead lighting does the same thing from a different direction — it removes the shadow that makes a jawline legible and puts dark circles under the eyes.
The practical move is to scan two or three photos taken in different conditions. The gap between your worst and best photo of the same face is usually larger than the gap between most people's faces — which is the useful, actionable version of the answer you came here for.
Fixable versus structural
Every reading is tagged. Some — grooming, brow shape, body composition, expression, how you angle your head — respond to things you can do this month. Others are bone, and no routine changes them. The report separates the two explicitly, because the expensive mistake is spending months on a reading that was never going to move.
Your next step
Score your own face free
Upload one selfie, get scored on 17 facial-geometry metrics + your validated Impression Percentile in ~30 seconds. No signup. Full written report $14.99.
Score my face free →Upload a selfie
Get scored, free
Unlock the report
Questions
Can a test actually tell me if I'm handsome?
Not as a verdict, no. Nothing measurable can tell you whether a specific person will find you handsome — that depends on context, personality, and chemistry a photo cannot hold. What this does is measure 17 facial proportions used in published attractiveness research (jaw angle, canthal tilt, facial-width-to-height ratio, facial thirds, midface ratio and others) and show where your photo lands against a large reference population. That is a measurement, not a judgment.
What does the score actually mean?
Your Face Score is a percentile: it says roughly where a still photo of your face sits relative to 5,500 human-rated faces from the SCUT-FBP5500 academic benchmark. It is cross-validated at about r ≈ 0.8 against how people actually rated those faces. It is a score for the PHOTO, not for you — the same face photographed differently can land in a different place, which is the single most useful thing the number tells you.
Which measurements matter most for men?
In the research literature, jaw and gonial angle, facial-width-to-height ratio, brow position and chin projection carry more weight in how male faces are read than they do for female faces. Canthal tilt and facial thirds matter across the board. Your result breaks all 17 out individually, so you can see which specific readings are doing the work rather than just receiving one number.
Is it free?
Yes. The scan returns your Face Score and your strongest reading with no signup and no email. The full written report — all 17 measurements, which are fixable versus structural, and a 30-day plan for your weakest one — is an optional one-time $14.99. No subscription.
I look better in the mirror than in photos. Which one is real?
Both, and the discrepancy is usually the camera rather than your face. A phone held at arm's length uses a wide-angle lens that enlarges the nose, flattens the midface and shortens the jaw — distortions the measurements pick up as proportion problems your actual face does not have. Mirrors also show you a flipped image you are habituated to. Shoot from about 4-6 feet under window light with a neutral expression and most people score noticeably higher on the same underlying face.
Does a beard, haircut or lighting change the result?
Yes, and that is the point rather than a flaw. The reading describes the photo you uploaded. Beard shape changes the measured jaw outline, hair changes where the upper third boundary falls, and lighting changes how much depth the camera records. Running two or three photos through and comparing is the most useful way to use this — it tells you which version of yourself the camera reads best, which is directly actionable for a dating profile or a work headshot.
How is this different from people rating me out of 10 online?
Voting sites average the opinions of whoever happens to be online, which is noisy and unrepeatable. This measures geometry, so the same photo returns the same number every time, and it tells you WHICH readings produced it. If you want the human version too, the Human Score feature has real people rate the same photo, and you can see the two verdicts side by side.
I got a low score. Should I do something drastic?
No. A low reading on a few proportions is not a medical finding and is not a reason to consider surgery. Most of the variance in casual photos comes from lens distance, lighting, angle and grooming — all reversible, all free. The report separates readings that are genuinely fixable from ones that are structural precisely so nobody spends money on the wrong thing. If you are weighing a cosmetic procedure for reasons unrelated to a photo, that is a conversation for a board-certified surgeon who can examine you in person.
Is the measurement biased toward one kind of face?
Partly, and it is worth naming. The reference ratings come from a large academic dataset, and any dataset of human ratings carries the preferences of the people who produced it. The tool is most reliable as a comparison — which of your own photos reads strongest under one consistent rubric — and least reliable as a statement about whether you are handsome in some absolute sense, which is not something geometry can settle.
What happens to my photo?
It is processed in memory and discarded. Nothing is stored, there is no account required, and no image is retained after the reading is produced.