Facial Thirds Test: Are My Facial Thirds Equal?
Upload one photo and get the upper, middle and lower thirds of your face scored out of 10 — forehead, nose and chin — alongside nine other features. Then see the finding the classical canon misses: across 17,032 analyses the three thirds are nowhere near equally important. Free — one tap to sign in.
Upload a clear photo
Click here or drag and drop your image
Choose one front-facing photo in even lighting
The three thirds, measured side by side
The face divides into three vertical bands: hairline to brow, brow to the base of the nose, and the base of the nose to the chin. The Renaissance canons said they should be equal. Nobody repeating that claim has ever checked whether the three bands carry the same weight in how a face is actually rated. We can check, because each third maps onto a feature we score — forehead for the upper, nose for the middle and chin for the lower.
| Third | Feature scored | Mean | Spread (SD) | 8+ faces | Under-5 faces | Gap | Score 7+ |
|---|---|---|---|---|---|---|---|
| Upper thirdHairline → brow | Forehead | 5.39 | 0.78 | 7.66 | 4.84 | 2.82 | 7.9% |
| Middle thirdBrow → base of nose | Nose | 5.36 | 0.93 | 8.01 | 4.51 | 3.5 | 7.7% |
| Lower thirdBase of nose → chin | Chin | 5.39 | 1.09 | 8.48 | 4.43 | 4.05 | 10.6% |
Read the mean column first and the canon looks vindicated: 5.39, 5.36 and 5.39 out of 10 — three numbers within 0.03 of each other. The average face really does score its three thirds almost identically, which is presumably why the equality claim has felt true for five hundred years. Then read the gap column, and it falls apart.
Figures use analyses from January 2026 onward (n=17,032), after the Dec 2025 model recalibration. Earlier analyses used a different scale.
Equal on average, unequal where it counts
The gap column is the distance between what the 560 faces rated 8 or above score on a feature and what the 4,307 faces rated under 5 score on the same feature. It is the cleanest available answer to the question "does this region separate people?".
The lower third separates them by 4.05 points — 5th of the 12 features, in the same territory as the jawline at 4.36 and not far behind the widest of all, jawline at 4.36. The upper third separates them by 2.82, which is 12th — last. That is a factor of 1.44× between two bands the classical canon treats as interchangeable.
The spread column says the same thing from the other direction. The forehead's standard deviation of 0.78 is the smallest of the 12: the whole population is squeezed into a narrow band around 5.39, leaving the upper third very little room to be either good or bad. The chin's 1.09 gives it genuine range. And the middle third has a quirk of its own — only 7.7% of people reach 7 or above on the nose, the lowest 7+ share of anything we measure, against 22.6% for the eyes. The middle third is the band where almost nobody excels, even though it correlates with the overall rating at 0.97.
So the honest conclusion is not "your thirds should be equal". It is this: equal thirds describe what a typical face looks like, because the three regions average within a fraction of a point of each other. They do not describe what a high-scoring face looks like, because the three regions contribute wildly different amounts to that outcome. A rule built on equality has no way to express that difference, and it is the difference that matters if you are trying to work out what to do about your face.
What this test does and does not give you
Worth stating plainly, because most pages on this topic imply a measurement they never produce. This tool scores the three regions as features out of 10. It does not output the height of each band in millimetres, and it does not output a ratio between them. If you want a single number for how closely your proportions track a classical ideal, that is the golden ratio test — which comes with its own honest caveat about how little that number predicts.
What you actually get
A forehead score covering the upper third — proportion, hairline shape, brow framing and the skin across it. A nose score covering the middle third. A chin score covering the lower third, read alongside the jawline. Plus facial harmony, which scores how the regions sit together rather than each one alone and correlates with the overall rating at 0.99, the closest of the 12. If what you wanted was a "do my thirds balance" answer, harmony is nearer to it than any ruler.
For context on where any of these scores sits: only 8.1% of people score 7 or above overall. A 6 in any of the three thirds is a better result than it feels like — and on the nose, where 7.7% clear 7, it is a considerably better one.
Getting an accurate reading of your thirds
Vertical proportion is the thing a photograph distorts most, and it distorts it in a predictable direction: whatever is nearest the lens gets bigger.
- Camera at eye level. Shooting from above enlarges the upper third and crushes the lower; shooting from below does the reverse. This one mistake accounts for most "my thirds are uneven" panics.
- Chin level, neither tucked nor lifted. Tilting the head changes measured lower-third height without changing your face at all.
- Hair off the forehead. A fringe does not shorten the upper third, but it does make it unreadable — and it will move the forehead score.
- Neutral expression. A smile shortens the apparent lower third and shifts the chin reading with it.
- Take three photos. Treat the range as your real result rather than trusting any single reading.
Frequently asked questions
What are the facial thirds?
A classical way of dividing the face into three vertical bands: hairline to brow (upper third), brow to the base of the nose (middle third), and the base of the nose to the bottom of the chin (lower third). The idea comes from the neoclassical canons of the Renaissance, which held that the three bands should be roughly equal in height. It began as a drawing convention and became a beauty claim.
Should facial thirds be equal?
Equal thirds are common, but common is not the same as high-scoring — and the three thirds are not equally consequential in the first place. Across 17,032 analyses the lower third, scored as the chin, separates faces rated 8+ from faces rated under 5 by 4.05 points. The upper third, scored as the forehead, separates the same two groups by only 2.82. The canon treats the bands as interchangeable. The measurements do not.
How do I measure my facial thirds?
With a straight-on photo, a ruler and four landmarks: the hairline, the brow line, the base of the nose and the bottom of the chin. Measure each band and compare them. Two warnings. A receding or uneven hairline makes the upper third impossible to measure with any precision, and camera height wrecks the result — shooting from above enlarges whatever is nearest the lens, which is your forehead. Shoot square, at eye level, with a neutral expression.
Does this tool give me a millimetre ratio between my thirds?
No, and it is worth being direct about that. The analysis scores the three regions as features out of 10 — forehead, nose and chin — alongside nine others, plus symmetry and an overall percentile. It does not output the height of each band or the ratio between them. If you want a single proportion-compliance number, that is the golden ratio test. If you want to know which region is actually holding your face back, the feature scores are the more useful answer.
What does an uneven lower third mean?
In this data, more than an uneven upper third does. The chin is someone's weakest feature 8.4% of the time and their best feature just 0.1% of the time — the 1st lowest best-feature rate of the 12 — so the lower third is a region that can cost you and almost never wins for you. It also has real range: 4.05 points between the high and low groups, against 2.82 for the forehead. The jawline, which sits in the same band, separates by 4.36. A weak lower third is a more meaningful finding than a weak upper third.
Why is the middle third so hard to score well on?
Because the nose has the lowest share of people reaching 7 or above of anything we measure: 7.7%, against 22.6% for the eyes. It still tracks the overall rating tightly at 0.97, so it is not that the nose does not matter — it is that very few faces score highly there. Being average on the nose is genuinely average, not a deficit.
Is a long lower third bad?
Not by itself. What the model reads across the lower third is chin projection and shape, jawline definition, and how the region balances against the two above it — not band height in isolation. Two faces with an identically long lower third can be a point apart on the chin score. The most useful thing you can do for a lower-third reading is photograph it honestly: chin level, camera at eye height, no upward tilt.
Is the facial thirds test free?
Yes. Forehead, nose and chin scores all come with the free rating, alongside nine other features, symmetry and your percentile. You sign in first — Google or Apple, one tap — and it is free.
Related tools
Rate My Face
All 13 scores, including all three thirds.
Golden Ratio Face Test
The proportion-compliance number, and what it is worth.
Face Symmetry Test
Horizontal balance, as distinct from vertical.
Forehead Rating
The upper third, and why it separates people least.
Nose Rating
The middle third, where almost nobody clears 7.
Chin Rating
The lower third, and the widest gap of the three.