Why the Same Size Fits Differently Across Brands

Lay three t-shirts on a bed (all labeled M, all from different brands) and measure them. Odds are good that no two will match: an inch of difference across the chest here, a shorter hem there, sleeves that end in different places on the same arm. This isn't sloppiness, and it isn't only the label inflation we've covered elsewhere. Two garments can carry identical tags, be made with identical care, and still be built for different bodies on purpose. Four mostly invisible decisions determine whether a garment works on you long before you touch it: who the brand fits its clothes on, how much room it designs in, how it scales between sizes, and how much its factories are allowed to vary.

Every brand fits its clothes on one specific person

Most clothing is developed on a fit model: a real person whose measurements match the brand's chosen ideal, and on whose body every sample is pinned, adjusted, and approved. From that process comes the brand's block, the base pattern from which everything else is derived. Buy from a brand and you are, in a real sense, borrowing its fit model's proportions: their shoulder slope, their torso length, the ratio of their waist to their hips.

Brands choose fit models to match the customers they imagine, and those imagined customers differ by market and heritage. A label developed for Japanese consumers, as Uniqlo originally was, tends toward trimmer shoulders and shorter sleeves than an American heritage brand cut for a broader frame, even after adapting its range for international stores. Neither block is wrong. But if your proportions sit far from a brand's fit model, no size in their run will ever feel quite right — you can change the tag, not the shape.

Ease: the room the designer meant to leave

Here is the concept that explains most "same size, different fit" confusion: ease, the gap between the garment's measurements and the body it's meant for. A shirt for a 100 cm chest is never 100 cm around, unless it's meant to be skin-tight. How much bigger it is depends entirely on intent. A boxy overshirt might carry 15 cm of ease; a slim dress shirt perhaps 8; a performance base layer can have negative ease, cut smaller than your body so the stretch fabric compresses against it.

Activewear makes this vivid. Leggings from Lululemon are engineered to measure smaller than you and hold on; a relaxed training tee from adidas is engineered to float away from you. Both can be "your size" while relating to your body in opposite ways. When a review says a garment "runs small," it often doesn't. It just carries less ease than the reviewer expected, which is a disagreement about intent, not an error.

Grading: sizes are extrapolations, not designs

Brands don't design every size. They perfect one sample size (often around a women's 8 or a men's M) and then grade it: add or subtract fixed increments at set points to generate the rest of the run. Grading assumes bodies scale uniformly, and bodies politely decline. Going up a size adds width everywhere by formula, but real people gain circumference in individual places; shoulder width and torso length barely change between many adult sizes, while grade rules keep adding to them anyway.

This is why fit complaints cluster at the ends of the size run: the sizes furthest from the sample carry the most accumulated extrapolation error. It's also why alpha sizing (S/M/L) feels blunter than numeric sizing: each letter typically spans what two numeric sizes would cover, so the grade steps are bigger and the "in-between" zones wider.

Tolerance: even the same garment isn't the same garment

Finally, manufacturing. Cut-and-sew production specs include a tolerance: an allowance, commonly on the order of a centimeter or two per point of measure, within which a garment still passes inspection. Fabric relaxes or shrinks between cutting and sewing, spreaders stack plies imperfectly, and two units of the identical style can land at opposite edges of tolerance. Stack a few of those against each other and one "medium" genuinely is bigger than its shelfmate.

Variance multiplies with the supply chain. A retailer aggregating hundreds of third-party labels, like ASOS, is really hundreds of blocks under one search bar. An ultra-fast-fashion marketplace sourcing from a vast rotating pool of suppliers, like Shein, can vary noticeably from listing to listing even within one nominal size. That is why per-item garment measurements, where published, are worth far more there than the size name.

So what do you actually do with this?

Once you see sizing as four stacked decisions (block, ease, grade, tolerance), the strategy writes itself. Stop asking "what size am I?" and start asking "what are this garment's measurements, and what is it trying to do on my body?" Concretely:

Know your own numbers cold. Chest, waist, hip, shoulder, inseam: our measuring guide takes about ten minutes and pays off on every purchase afterward. Prefer garment measurements over body size charts whenever a retailer publishes them, and compare against a similar garment you own that fits well. Learn each brand's block once. A brand's fit is far more consistent with itself than with anyone else, so one successful purchase teaches you the whole label. And read fabric content before reviews: a few percent of elastane changes how much ease a garment needs more than a whole size step does.

The label was always the least informative thing about a garment. The block it came from, the room it was designed to leave, and the tolerance it shipped within tell you whether it will fit. Those are knowable (measurable, even), and once you start reading them, the three mismatched mediums on the bed stop being a mystery and start being data.