'How attractive am I?' What a facial analysis can and can't tell you

It is one of the most quietly typed questions on the internet, and it deserves a real answer instead of a number with a confidence it has not earned. Here is what a facial analysis can honestly tell you, and what it genuinely cannot.

What it can measure

It can measure your proportions — thirds, symmetry, eye spacing and tilt, jaw and nose ratios — and compare them to documented patterns and to a lot of other faces. Done properly, the headline is not an average of those ratios; it is a model trained on thousands of human ratings and calibrated to read as an approximate percentile within your sex. In our testing that model agrees with human raters about 88% of the time for women and 87% for men, and it stays fair across ethnicities on faces it never saw in training. So it can give you a reasonable sense of where your proportions sit relative to others. That is real information.

What it cannot

It cannot tell you whether a specific person will find you attractive, because attraction is personal and situational. It cannot see expression, charisma, voice, humour, style, grooming or the way you carry yourself — and in real life those move attraction enormously. A neutral photo deliberately strips most of them out. And it cannot measure your worth, which is not a proportion.

Why one number is the wrong lens

Attractiveness is not one-dimensional, and "rate my face" culture quietly corrodes self-image, especially for younger people. A proportion score is a snapshot of geometry on one day, under one light. The people who read as most attractive are rarely the ones with the most textbook measurements; they are usually the ones who look after themselves and seem at ease.

If you are going to look, look like this: treat the result as neutral information, notice the genuinely easy wins (grooming, fitness, sleep, better photos), learn which "ideals" are actually evidence-based — and then close the tab. It is a guide, not a verdict.