Insights

Not wider. Elsewhere.

When a shoe hurts across the front of the foot, the advice is to go up a size. It rarely helps, because the problem is not length but a place: the joint where the shoe turns inward and the foot does not.

October 2026 · Novainvent AB, Halmstad · 7 min read

Part 2 of a series on whose foot a shoe is built for. Part 1: Nobody has the ideal foot. That is why we built one.

When a shoe hurts across the front of the foot, the advice is always the same: go up a size. Sometimes it helps. Often it does not, and the reason is geometric rather than mysterious. The foot is not too long for the shoe. It is the wrong shape for it, at one specific place, and length has very little to do with that place.

Where the foot is widest

The widest point of a foot is not the toes. It is the joint where the big toe meets the foot, the first metatarsophalangeal joint, and its counterpart on the outside, the fifth. Between them runs the ball of the foot. Everything in front of that line, the toes, is narrower and more flexible. Everything behind it tapers towards the heel.

A shoe is built around a last, and a last has the same widest point. The question is what happens in front of it. On many lasts, the inner line begins to turn inward almost immediately after the ball: the toe box narrows towards a point somewhere in front of the second toe. On a foot with a straight big toe, this costs little. On a foot whose big toe already angles outward, the shoe pushes exactly where the joint is trying to go the other way.

That angle has a name, the hallux valgus angle, and it is the most important number in this article, because it is not a number. It is a distribution.

What the same size hides

A size has one last, and a last has one toe box. It is drawn for a foot with a straight big toe, because that is what the ideal foot has. The question is how many of the feet that will wear the shoe actually look like that.

The most thorough answer comes from a meta-analysis of 76 population surveys covering nearly half a million people. It puts hallux valgus, defined as a big-toe angle above the clinical threshold of 15 degrees, at 23 per cent of adults between 18 and 65 and at 36 per cent over 65. Among women it is about 30 per cent, more than twice the rate among men. A newer global meta-analysis, pooling 45 studies, lands a little lower, at 18 per cent for Europe and 24 per cent for women, with the same twofold gap between the sexes. Between one in five and one in four, then, depending on who is counted and how.

And the threshold is not a wall. When Hardy and Clapham measured the joint on X-rays in 1951, the feet of people with no complaint averaged 15.5 degrees and the feet of those waiting for surgery averaged 32. Between a straight toe and a surgical one lies a continuous spread, and most feet sit somewhere along it rather than at either end.

Read that against how shoes are made. One last, one toe box, one straight toe, for a population in which the most decisive forefoot dimension runs from a few degrees to more than thirty, and in which every fourth foot sits beyond what the last was drawn for.

The careful sentence

It is tempting to write that shoes cause this. The evidence does not quite support that sentence, and it is worth being precise. Heredity weighs heavily: the shape of the joint and the ligaments around it are largely inherited, and the condition exists, at low rates, in people who have never worn shoes. What the evidence does support is narrower and more useful.

On the island of St Helena in the early 1960s, a medical officer examined some 4,000 islanders, many of whom had grown up barefoot. Among those who had never worn shoes, about two per cent had hallux valgus. Among women who had worn shoes for sixty years, nearly half did, and the rate rose with every year of shoe wearing. Fifty years later, a cohort of 2,627 British women aged 50 to 89 showed the same thing from the other end: those who had worn shoes with a very narrow toe box in their twenties were almost three times as likely to have hallux valgus in later life. Pooling ten such studies, a 2026 meta-analysis puts the odds of hallux valgus at 2.25 times higher for narrow-toed shoes than for wide ones, and 1.65 times higher for high heels. The most recent experimental work, from the same Australian group behind the fit review below, treats narrow footwear as a modifiable risk factor and measures exactly where the pressure lands: on the inner side of the forefoot, at the joint.

So the sentence that holds is this: a shoe designed for a foot you do not have will hurt, and the place it hurts is not the end of the toes but the joint behind them. Which is why the usual remedy fails.

Why a larger size does not help

Go up a size and two things move. The shoe becomes longer, and the widest point of the last moves forward with it, by six or seven millimetres per European size. The foot does not follow. Its joint stays where it was.

The shoe is now too long at the front and no wider where it needed to be. This is the failure that width systems were invented to address, and it is also why they are so rarely used: a width letter tells you the girth at the ball, not where the inner line turns or how it treats a toe that angles outward. Two lasts with identical width can be entirely different shoes for the same foot.

The widest study of fit we have, a systematic review of eighteen studies, found that between 63 and 72 per cent of people wear shoes that do not match their feet in width or length. Width was the dominant failure. Almost nobody is walking around in shoes that are simply too short.

What can be measured instead

If the problem is a place rather than a length, the answer is to compare feet and shoes at that place. Where is the widest point of this last, and where is the joint of this foot? Where does the inner line of the last begin to turn, and what angle does this foot's big toe already have? These are questions about geometry on both sides, and both sides can now be measured: the foot from a phone, the shoe from the last it was built on.

Asked that way, the answer is rarely "a size up". More often it is: this shoe, in your usual size, will press here; that shoe, from a different last, will not. The comparison happens at the ball and along the inner line, zone by zone, and it treats the toe angle as a shape to be accommodated rather than a defect to be corrected.

It also gives designers something they have never had. Part 1 of this series described the ideal foot: the statistical centre of a population, the foot a last is drawn for. Its big toe is straight. That is the zero point for this angle, and it is now possible to say how far a given toe box serves from that zero point, and what share of real customers it leaves outside. One in four is a large share to leave outside by accident.

The bunion is not a problem to be fitted around by going up a size. It is a measurement, and it belongs in the comparison.