Oblong face shape: the 1.5:1 rule almost never fires
Noticeably longer than it is wide, with forehead, cheekbones and jaw close to the same width down the whole face. Almost every description of it comes with a number attached — a face about one and a half times longer than it is wide. We have measured 43 faces with the detector this site ships. One of them reaches 1.5.
The distribution behind the rule
| Shortest face measured | 1.18 |
|---|---|
| Median of all 43 | 1.36 |
| Median of the faces classified oblong | 1.48 |
| Longest face measured | 1.53 |
| Faces reaching 1.50 | 1 of 43 |
A threshold that one face in 43 clears is not separating a category from its neighbours, it is naming an extreme. Which is fine as description and misleading as a test — and it is quoted as a test constantly, including by tools that then return oblong far more often than one in 43, which means they are not applying the number they cite.
Now the part that cuts against us, because this whole argument is a ratio and the top of that ratio is an estimate. The face mesh has no hairline landmark: the topmost point sits on the upper forehead, below the hairline. Full face length is the mesh length multiplied by a constant, 1.12, and our own source calls it what it is — a factor that still needs calibration against labelled photographs. Choose a larger constant and every number in the table above moves up. The shape of the distribution survives that, the absolute values do not, and anyone quoting our 1.53 without quoting the constant is quoting half a measurement.
Now measure your own
Reading a description and holding your own face up against it is guesswork, and the descriptions overlap on purpose, because real faces do. The tool measures the four lengths in the diagram plus the jaw angle, in your browser, and shows you how close you sit to every shape rather than picking one and hiding the rest.
The length in this result is a mesh measurement scaled by the uncalibrated constant named in the table above, so treat the ratio as comparable between faces rather than as an absolute number.
Front-facing photo, even light, hair off the forehead
Your four measurements will appear here, so you can see which shape they sit closest to.
What our own classifier does with oblong
Four of the 43 faces came back oblong, the second fewest of the six. When oblong wins, the runner-up is oval three times out of four, which is exactly the pair you would expect: both are long, and they differ by whether the sides run parallel or curve.
Length against width is the reason oblong is ruled out for 29 of the 43 faces. That same measurement is also the reason round is ruled out 31 times — for the opposite reason. One ratio runs both shapes from opposite ends: oblong is refused for being too short, round for being too long, and everything in between falls to the residual. It is the cleanest illustration on this site of why six labels on four measurements leaves so much to the thresholds.
Everything is in the middle of its range. Oval is the shape defined by not being any of the others. That is the neighbour oblong is competing with for most faces, and the full grid of which shapes get confused with which is on the face shape chart.
Oblong face shape questions
What is an oblong face shape?
+
Noticeably longer than it is wide, with the forehead, cheekbones and jaw close to the same width the whole way down. The sides run almost parallel. Length is doing the work rather than any one width, which is what separates it from every other shape on the list.
Is an oblong face 1.5 times longer than it is wide?
+
That is the rule in circulation, and in our measurements it almost never fires. Across 43 distinct faces the median length-to-width is 1.36, the longest face we have measured is 1.53, and exactly one face reaches 1.5. If the rule were the definition, oblong would be a category with one member. It is better read as a description of an extreme than as a threshold anything is actually tested against.
What is the difference between an oblong and an oval face?
+
Length, and then the sides. Both are longer than they are wide. An oblong face runs close to parallel from forehead to jaw, with the three widths similar all the way down; an oval face is widest at the cheekbones and curves inward above and below them. This is the one confusable pair with a real anthropometric measurement behind it — the morphological facial index, face height over cheekbone width — but that scale is five bands on one continuum and no publication maps it onto the styling pair, so it will describe how long a face is without telling you where oval stops.
Is an oblong face shape rare?
+
In our results, yes: 4 of 43 distinct faces came back oblong, the second fewest of the six. But that is a fact about where our thresholds sit rather than about how common oblong faces are, and given how rarely the 1.5 rule fires it is also a sign that thresholds inherited from that rule will under-report the shape.