which measurement do i actually go by.

Published August 18, 2026·File 11 · Sizing·10 min read

Your shoulders say one size, your waist says another, and the chart has no row for the measurement you actually care about. Here is how size charts are built, what sizing up really buys you, and the rule that decides it — including where that rule flips on our own tees.

You have the tape measure out. You have the chart open in another tab. Chest says medium, waist says small, and the thing you are actually worried about — whether the shoulders will pull the second you reach for anything — is not on the chart at all. So you do what everyone does. You size up, it arrives, and now the shoulders are almost right and the rest of it hangs on you like a bin bag.

Someone on a menswear forum put the whole problem in one sentence back in 2015, and nobody has improved on it since: "Larges fit fine around the abdomen but not around the shoulders, and XLs fit fine around the shoulders but are too frumpy in the waist." Frumpy is the giveaway. The failure is not that the garment hurts. It is that it looks wrong, and you cannot point at which number caused it.

First: which chart are you actually looking at

Two documents wear the same name and answer different questions. A body chart describes you. The industry standards behind most of them are explicit about it — ASTM D5585 states that its tables "list body measurements of adult female misses figure Type sizes 00 through 20", and the men's equivalent, D6240, opens the same way. Our own chart is one of these: it tells you to wrap a tape around the fullest part of your chest.

A garment chart describes the thing you are buying, lying flat on a table. A typical streetwear blank publishes exactly two points of measure — body width and body length — and body width is usually half the chest, so you double it before comparing it to anything.

The gap between the two has a name. "Ease is the difference between the finished garment measurements and your actual body measurements." It splits in two: wearing ease, the room you need to move and breathe, and design ease, the extra the designer added because the look requires it. A boxy tee is not accidentally large. It is carrying design ease on purpose.

And ease is why a knit behaves differently from a woven. On a woven, the garment has to be bigger than you. On a knit, "the garment is actually smaller than your body. This works because the fabric stretches to fit you." Negative ease is normal in a rib tank and impossible in a woven shirt, which is most of why the same body needs different logic in each.

What sizing up actually buys you

Here is the part that explains the bin-bag problem. Grading is how one sample size becomes a whole size run, and the growth is not spread evenly. Working from published grade rules, each size up adds, on the half-front of the pattern: 3mm at the neckline, 3mm along the shoulder line, and 6mm at the side body. That totals 24mm on the half front, mirrored on the back, which is where the familiar two inches per size comes from.

Do the division and the whole frustration resolves. Half of everything a size adds goes into the side seam. A quarter goes to the shoulder line. A quarter goes to the neckline. You are not buying shoulder room when you size up. You are buying circumference, with a rounding error of shoulder attached.

This is also why proportional grading cannot reach a body whose shoulders and waist disagree. The grade is a fixed ratio applied to every size in the run. It has no mechanism to move one measurement independently of the others. Going up the run changes all three at once, in the same locked proportion, forever. If your ratio is not the standard ratio, there is no size on the run that fixes it — which is a design limitation, not a fact about your body.

We ran 6,068 bodies through our own size chart

Rather than argue about this, we tested it on ourselves. ANSUR II is a public anthropometric survey of 6,068 adults — 4,082 men and 1,986 women — measured by trained anthropometrists across 93 dimensions. We took every one of those bodies and assigned it a size using our own published chest and waist bands, then asked a simple question: does the chart give each body one answer or two?

It gives two. Of the bodies that land on our chart in both columns, 65.9 percent are told one size by their chest and a different size by their waist. Only about a third get a clean single answer. And 34.6 percent of the survey falls off the chart entirely at one end or the other, which is its own finding.

The other number worth sitting with is the spread inside one size. Take a single two-inch chest band — one size's worth — and the waists inside it range across about seven inches. At a two-inch grade, that is three and a half sizes of waist hiding inside one size of chest. The chart is not being careless. It is doing the only thing a two-column chart can do.

The rule, and the one place it flips

For a structured woven with a seam sitting on your shoulder point, the shoulder governs and everything else is negotiable — because the shoulder is the one thing money cannot easily fix. One tailoring guide is blunt about the cost: "Making significant changes to shoulder width usually requires completely rebuilding the jacket, and can cost $200-400 or more", against thirty to fifty dollars for waist work. Another says shoulders are "nearly impossible to alter once constructed." Both are talking about jackets, and nobody prices a shoulder rebuild on a tee because nobody does it — but the asymmetry is the argument.

Then there is the honest part, which cuts against the tidy version of this advice. Our own tees are drop-shoulder. The seam does not sit on your shoulder point, so it is not a boundary, and the shoulder-governs rule does not transfer. On a knit or a dropped shoulder, the fabric and the drape absorb the miss, which is exactly why our size guide tells you to size down between sizes rather than up, and to go up only if the chest binds. Advice that ignored our own construction would be tidier and worse.

---

Quick answers

Is the size chart showing my body or the shirt?

Check the instructions. If it tells you to wrap a tape around your chest, it is a body chart and the finished garment will be larger than those numbers — that gap is called ease. If it lists body width or says laid flat, it is the garment itself. Scared But Alive publishes a body chart. Most streetwear blanks publish garment charts.

My shoulders, chest and waist are three different sizes. Which one wins?

On a structured woven with a seam sitting at your shoulder point, the shoulder wins, because it is the one thing a tailor effectively cannot fix. On a knit or a drop-shoulder cut the shoulder seam is not a boundary at all, so buy to the chest and let the drape carry the rest.

Will sizing up fix tight shoulders?

Barely. In standard grading a size adds about two inches around the body, but only an eighth of an inch to the shoulder line of each pattern piece. Half of the new fabric goes into the side seam. You will gain a loose waist long before you gain meaningful shoulder room.

Why doesn't the size chart have a shoulder measurement?

Because almost none do. The industry body-measurement standards, ASTM D5585 for women and ASTM D6240 for men, are built around chest, waist and hip. Our own chart lists chest, waist and hips. A typical blank tee chart lists body width and body length. Neither format has a shoulder row.

Is it normal that the chart gives me two different sizes?

It is the majority case. We ran 6,068 adult bodies from the public ANSUR II survey through our own published chest and waist bands. Of the bodies that landed on the chart at all, 65.9 percent were told one size by their chest and a different size by their waist. Roughly one in five disagreed by two sizes or more.

Can a tailor fix shoulders that are too tight?

Mostly no. One tailoring guide calls shoulders nearly impossible to alter once constructed and budgets 75 to 150 dollars if attempted. Another puts a real shoulder rebuild at 200 to 400 dollars or more, against 30 to 50 for waist work. Both are jacket figures, but the asymmetry is the point.

Keep reading

• what's my size, then. — File 02: why the three-size body is the majority, not the exception
• why does the armhole always show everything. — File 01: the geometry the shoulder line hands off to
• why don't men's shirts ever fit me right. — File 10: the block underneath all of this, and the four-way miss
• why does every boxy tee tent on me. — File 06: what design ease is supposed to do when it works

---

References

A note on our own numbers. The 65.9 percent figure is our arithmetic, not a published study: we took the public ANSUR II measurements and scored each body against the chest and waist bands on our own live size guide. ANSUR II surveyed U.S. Army personnel, who are leaner and younger than the general public, so the disagreement rate in a civilian population would likely be higher, not lower — treat it as a floor. Trained anthropometrists also measure differently from a soft tape in a bathroom. The grade rules are one experienced pattern cutter's teaching figures rather than a legal standard, and the alteration prices are for tailored jackets. Product education from the development room.

feel scared. feel so alive.