What the model does and what it cannot know

This service estimates how sustained stretching during storage affects the elasticity of a sock cuff. The estimate rests on three things: the percentage of elastic fibre printed on the composition label, the rib structure and band width read from your photograph, and the storage duration you enter.

This is not a test of the specific pair.

Nothing is measured physically. No sample is taken, no tension is applied, and no instrument touches your socks. What comes back is a reading of the three inputs above against a model of how elastomeric yarn behaves when it is held at a fixed extension.

An image model is used for one job: reading the elastic percentage from a composition label, and reading the rib structure and band width from a photograph of a cuff. That is the whole of its involvement.

The model does not estimate slackening. The estimate itself is computed in your browser, from the values above, and it never leaves your device.

Any wording in a reading is generated text and can be wrong in the way generated text is wrong. The figures beside it are arithmetic, and they are arithmetic you can check against the inputs shown next to them.

A composition label gives only the percentage of elastic fibre. It never gives its type or its recovery grade.

Different elastic fibres differ considerably in how well they resist static fatigue — in how much of a held extension they give back. Two socks both marked 3% can behave quite differently, and no photograph of a label will tell you which is which. This service cannot read that layer, and neither can you.

Two further things are unknown and stay unknown: the temperature and humidity of the drawer, and how long the pair has already been in service. Both change the answer.

For those reasons all estimates are given as ranges, and no assurance is given that any storage method prevents slackening. Nothing here should be read as a promise about the future condition of a garment.

In most cases the effect is small. For socks with a higher elastic content, a wider rib and frequent use, storage matters far less than wearing and washing.

The service says so plainly. Every reading includes a section naming the pairs that need no attention at all, and that section is never behind a paywall and is never obscured by a watermark. A tool that only ever tells you that you are doing something wrong is manufacturing worry, and this one is built not to.

The narrow case where storage does add up is a band with little or no elastic fibre, a narrow rib, and months of sitting rolled. That case is real, and the reading will tell you when you are in it.

This service recommends no storage product, no sock brand and no merchandise of any kind, and it provides no purchasing route.

Any method described is generic — laying a pair flat, folding once, leaving the band alone — and points at no product. There is nothing to buy here beyond the subscription, and nothing is sold on anyone else’s behalf.

High-risk uses are prohibited. Readings must not be relied on for medical, legal, financial, safety, employment, credit, housing or judicial decisions.

The service does not accept photographs of socks being worn, and nothing it produces contains a person, a body part or a pair of hands. See the Acceptable Use Policy.

Photographs you upload are not used to train, fine-tune or evaluate models — not ours and not any third party’s. Retention periods and the deletion path are set out in the Privacy Policy.