Your Face Shape: Round

Your Round Result: FAQ

How much weight should the confidence figure on a round result carry?

It says how far your five measurements sit from the round row of the published prototype table, relative to how far apart the seven rows are from each other. It is a distance, not a probability, and it is not an accuracy claim — this site publishes no accuracy figure on any page, because we could not verify one. On a round result there is a particular reason to read it closely. Round and square are the two rows this site's rule set enters most similarly: the three widths run much the same way on both, so nearly all of the distance between them sits on one reading, the corner at the jaw. That is also the weakest of the five in this site's own calibration photographs, and on the calculator path it is estimated from your four widths rather than measured — which is why the detector scores all five at once and the page shows a margin instead of a verdict.

Which row of the measurement table decides a round result?

That is the round row of the prototype table this site classifies against, plus or minus one standard deviation, converted onto the tape-measure scale so it lines up with the numbers you actually produced. Read it to find which feature decided the result and which were unremarkable; on a round result the jaw angle row is the one to look at first. Worth knowing about that column: the spread beside it is the same width on all seven result pages, because it comes from a single per-feature figure rather than one fitted per shape. A generous band on the jaw angle row is therefore a statement about how much faces vary, not about round being a loose category. The round row itself was fitted to measured group means from our reference set.

What does it mean when the square bar runs nearly level with round?

Each bar is the same distance measure as the headline figure, drawn once per shape, so you can see whether round won outright or narrowly. They are not a ranking of faces and they are not the proportion of people with each shape — no prevalence figure appears anywhere on this site, because the ones in circulation come with no stated sample and no stated method. On a round result two bars deserve your attention. Square is the nearest boundary in the rule set, and the difference between the two rows is entered on the corner at the jaw — the weakest of the five readings in this site's own photographs, so it is a margin to read, not a line a face has to clear. Oval sits almost exactly as close behind square and rests on different readings, so the two bars tell you different things: a high square bar means the angle at your jaw falls between the two rows, while a high oval bar means your face length does.

My result says round with no second shape attached. Does that make it certain?

No, and the absence is worth explaining. A second shape is named on the result only when the runner-up sits close enough to the winner to be worth flagging — the site would rather show you a borderline case than bury it. No second shape means the gap was too wide to flag, not that the measurement was exact. Two things still apply. The confidence figure can be modest on a clean single-shape result, because it measures distance to the prototype rather than margin over the runner-up. And the seven bars name the runner-up whether or not it was close enough to be flagged. On round the boundary to watch is square, decided by the jaw angle, so read the square guide too even though this page did not flag it.

The calculator said round and the photo said square. Which is right?

That is the most informative disagreement this site can show you, and the reason is specific to this pair. Round and square are the pair the rule set enters most similarly, differing on the jaw angle more than on anything else — and that is the weakest of the five readings on this site's own photographs, and the calculator never measures it. It estimates it from the four widths you typed, then lowers the confidence ceiling to reflect that. So when the two paths land on opposite sides of this boundary, the photo path is the one that actually looked at the angle. Re-run it with your hair clear of your jaw and your head level. If it still says square, read the square guide's version of this comparison before deciding. The disagreement is printed near the top of the page rather than hidden, for exactly this reason.

Should I change my haircut because the result says round?

Not on this result alone. The picks below are a styling convention applied to a measurement, and the measurement that separates round from square is the one feature a photograph can read and a tape measure cannot. If the confidence is unremarkable and the paths agree, the picks are as good a starting point as any guide gives you. If they disagree, or the square bar runs level with the round one, read both sets of picks — they overlap considerably, because both shapes carry a soft width and differ mainly at the corner of the jaw. And nothing is being sold to you here: the neutrality note above the picks is the whole of the commercial arrangement.