Shoes Records: What the Sizing Evidence Says

Sizing arguments are vocabulary arguments before they are measurement arguments. Two buyers can hold the same pair, read the same listing and still disagree completely, because one is talking about the length of the insole while the other is talking about the length of a foot. The twelve terms below are the ones that decide those disagreements, and footwear is where they carry the most money: the Shoes category holds 24 records priced from $24.09 to $100.84. Each term is given with the meaning it carries, the misuse it attracts, and the usage that survives contact with a delivered parcel.

The order is alphabetical, running from batch variation through to US Women. Nothing in the list is specific to one model or one brand: these terms describe how sizing is stated, claimed and measured, which is the part that transfers between records.

A term list does not settle a fit question. It only makes the question askable, and it separates two statements that are usually blended together in a listing description. Once the vocabulary is fixed, a listing can be read against one term at a time instead of against the whole of it.

The index below holds the same twelve terms in the order they tend to appear inside a purchase decision: what the listing says, what the label means, what the pair measures, and what the buyer does about the difference. Each line points at the term that does the work in that stage.

1. Seller size chart: the claim, read first and remembered as a claim.
2. EU sizing and UK sizing: two label scales, both converted through the chart rather than by rule of thumb.
3. US Men and US Women: the same label question in a second scale, and the place where unisex listings most often go wrong.
4. True to size: a model-level claim that means nothing until a reference pair is named.
5. Batch variation: the reason two pairs of one model can measure differently.
6. Last width: the dimension that a length-only conversion silently drops.
7. Insole length and outsole versus insole: the two measurement terms, and the pair most often confused.
8. Measurement versus claim: the labelling rule that keeps the previous two terms honest.
9. Half size up: the adjustment, applied deliberately and recorded as an adjustment.

Photo coverage in the pool makes several of these terms checkable before shipping. Of the 195 records extracted on 2026-09-29, 158 carry five or more images, a share of 81 percent, and the median record carries 14 images against a maximum of 46. A set of that size will usually include a frame of the label inside the collar, which is where batch variation stops being theoretical and becomes a printed string of characters.

The index is meant to be read twice. The first pass tells you which terms the listing addresses. The second pass tells you which terms it avoids, and the avoided ones are where the fit risk sits.

The conversion block at /instrument/size-evidence/ begins from the standard chart rather than from the pool. That chart covers EU 36 through 47 against US Men 4 to 12.5, US Women 5.5 to 14, UK 3.5 to 12, and a foot length range of 22.5 cm to 29.5 cm. It answers one narrow question well: what does a printed label mean in another scale.

What the instrument does not do is publish measured evidence, and that restraint is deliberate. The measured insole table behind /instrument/size-evidence/ is empty. A row appears there only when the site holds a measured insole length, a sample count and a source type at the same time, and no brand has reached that bar.

An empty table states more than a table of estimates would. A filled table built from invented numbers would be read as evidence and then used to choose a size, which is the single decision it would get wrong at the expense of the person reading it. The same rule governs the reference tables on this site: a figure without a traceable origin does not enter a data file.

The photo records are where the question gets a partial answer instead. Six shoe records in the pool show what can be checked before a parcel leaves: /middlemen/nike-dunk-low-sneakers-2024-new-color-skateboard-4780232366/, /middlemen/golden-goose-chunky-sneakers-23-styles-8892968761/, /middlemen/gucci-shoes-36-style-0207817317/, /middlemen/new-balance-550-sneakers-33-style-9441148506/, /middlemen/dior-b22-sneakers-8063020218/ and /middlemen/prada-chunky-sneakers-6-styles-2634374850/. Those six sit inside a category of 24 records whose median price of $68.68 is the highest of the eight categories in the pool.

Inside a photo set, the frames that carry sizing information are the ones showing the label inside the collar and the insole if it has been pulled out. A picture of the outside of a shoe confirms the model and says nothing about fit. For a category whose median record costs $68.68, a legible label frame is the cheapest evidence a buyer can obtain before committing.

The order of operations that follows is short. Read the term, read the label, read the chart, then ask whether a measurement exists. Where it does not, the honest position is that the fit is unknown, and the chart row closest to a measured foot length is the least unsupported choice available.

Three quantities settle most of these arguments before they begin. The seller label is what the listing prints. The source chart is what the seller publishes about that label. The measured length is what one pair actually came to when it was laid flat. Only the third is a measurement, and it is the only one that cannot be revised after the fact. A record that carries a label and nothing else is telling you which of the three you hold, and that is more useful than a fit verdict would be.

The order of operations that follows is short. Read the term, read the label, read the chart, then ask whether a measurement exists. Where it does not exist, the honest position is that the fit is unknown, and the chart row closest to a measured foot length is the least unsupported choice on the table.