How We Calculate Shoe Sizes

We do not pretend a single true size exists. We combine deterministic conversion formulas with measured brand and model fit data, then publish a best fit with a likely range, confidence score, and source trail.

The two-layer model

Layer A: formulas. US, UK, EU, and CM systems are converted through one stable axis: foot length in millimetres.
Layer B: empirical fit. Brand and model data adds measured last and insole signals plus fit offsets for shoes that run small or large.

Confidence score

Each recommendation receives a 0 to 1 confidence score based on source count, lab measurements, sample size, and source disagreement.

When sources disagree by more than half a size, confidence is reduced and the range is shown instead of hiding the uncertainty.

Best fit vs strict conversion

Strict conversion is the formula result, such as US to UK or EU. Best fit is the size we would actually recommend after brand and model fit data is applied.

Because even a few millimetres can change comfort, we show a likely range rather than a falsely precise single number.

Keeping data current

Official charts are re-checked regularly.
Every sourced record keeps a last verified date.
Low-confidence records remain visible but clearly labelled.