T-shirt sizing, settled.

01 — How to measure

How to measure a shirt (the only method that works)

Nobody knows their t-shirt size. They know what fits them, which is a different thing — a Medium in one brand is a Small in another, and the number on the tag has never been standardized. So skip the guessing and measure a garment.

Find a shirt the person already likes wearing. Lay it flat on a table, front side up. Smooth the wrinkles out but do not stretch it. Then take two measurements:

• Chest width — measure straight across from one armpit seam to the other. This is a half measurement. Every chart on this page uses it, because it is the number you can actually get with a tape and a table.
• Body length — measure from the highest point of the shoulder seam, right where it meets the collar, straight down to the bottom hem.

Match those two numbers against the chart for the specific style being printed. If the person lands between two sizes, size up — a slightly roomy shirt gets worn, a tight one does not. That single rule prevents more reorders than anything else in this guide.

02 — Size charts

Size charts for the blanks we print

These are the styles that cover the overwhelming majority of what leaves our shop. All chest widths are laid-flat half measurements. All figures are manufacturer specs; garments carry a tolerance of roughly ±1 inch, and garment-dyed styles like Comfort Colors run closer to ±2 inches by design.

03 — Brand fit

How the brands actually fit next to each other

Four Larges, four different shirts. Same chest width on the tape, very different experience on a body — because weight, cut and length do most of the work:

Relative roominess at the same nominal size. Not a measurement — a judgment call from printing these blanks week in and week out.

The practical takeaways:

• Switching a group from Gildan to Bella+Canvas without changing the size breakdown is how you end up with complaints. Shift roughly 5–8% of the run one size up.
• Going the other direction — Bella to Comfort Colors — you can hold the breakdown steady, but warn people about shrinkage.
• Never mix two brands in one run and print the same size grid on both. Size each brand off its own chart.

04 — Group orders

Sizing a group order when you can't ask everybody

The ideal is always a sign-up sheet with real sizes on it. In the real world you get "about 200 shirts, mostly adults, we need them Thursday." Here is how to think about it.

Start from a base grid for a general adult audience, then move percentage points around based on what you know about the group. The base:

Weighted toward L and XL on purpose. American adult sizing has moved up over the last two decades, and an XL that goes unclaimed still gets worn by somebody — an unclaimed Small usually doesn't.

The math is simple enough to do on a napkin: quantity × percentage ÷ 100. For 250 shirts at 27% Large, that's 250 × 27 ÷ 100 = 67.5, round to 68. Round the big middle sizes up and the tails down, and make the totals reconcile.

05 — Calculator

Size breakdown calculator

Enter a quantity, pick the closest audience, and this will do the arithmetic. Treat the output as a starting point to argue with, not a verdict.

06 — By event type

Sizing by event type

Elementary schools

Almost entirely youth sizes, with a handful of adult shirts for teachers and volunteers. A workable split for a whole-school order: Youth S 15%, Youth M 28%, Youth L 32%, Youth XL 15%, Adult S–L 10%. If the order is for a single grade, ask the teacher — they know their room, and a fourth-grade class skews very differently from a kindergarten class.

Middle and high schools

The awkward middle. Youth XL and Adult S overlap heavily, and teenagers overwhelmingly prefer the adult tag. Order Youth L/XL sparingly and lean adult: Adult S 15%, M 25%, L 25%, XL 18%, 2XL 10%, 3XL 7%, plus a few Youth L/XL for the smallest kids.

College events and Greek life

Runs smaller and younger than a general adult crowd, and oversized is the prevailing style — meaning people order up on purpose. For a women-heavy event: S 22%, M 32%, L 26%, XL 13%, 2XL 5%, 3XL 2%. For a men-heavy event: S 5%, M 18%, L 32%, XL 27%, 2XL 12%, 3XL 6%.

Corporate staff and office teams

You usually have a roster, so collect real sizes — a two-question form beats any percentage grid. When you can't, use the base grid and add one extra 3XL for every 50 shirts. Nothing sours a company giveaway faster than the largest person on the team not having a shirt.

Trades, construction, warehouse crews

Skew heavily large and buy up the tail: M 10%, L 25%, XL 30%, 2XL 22%, 3XL 10%, 4XL 3%. Also worth specifying a heavier blank — a 4.2 oz retail tee will not survive a job site.

5Ks and endurance events

Registration forms capture sizes, so use them — but expect a smaller-than-average distribution and a heavy women's presence in most road races. Order technical or lightweight blanks and keep a stack of L and XL at packet pickup for the day-of registrants.

Retail merch drops

Different game entirely: you are not covering a known group, you are managing inventory risk. Go light on the tails and deep on the middle — S 12%, M 25%, L 28%, XL 22%, 2XL 9%, 3XL 4% — and reprint the sizes that sell rather than guessing wider up front.

07 — What to avoid

Five mistakes that cost real money

• Ordering exactly your headcount. Every group order picks up people who weren't on the list. Add 5–10% and weight it to L and XL — it is far cheaper than a second run of eight shirts.
• Copying last year's breakdown onto a different blank. Same grid, different brand, different fit. Size off the chart for the style you're actually printing.
• Assuming women want women's cuts. Many prefer unisex. Offer both, or ask. Don't decide for them.
• Forgetting the top of the range. A 3XL costs a little more per piece and covers the person most likely to be left out. Order a few even when nobody requested one.
• Ignoring shrinkage on garment-dyed shirts. Comfort Colors and similar styles lose real length in a hot dryer. Either size up or put wash instructions in the box.

08 — FAQ

Frequently asked questions

Measurements above are drawn from current manufacturer spec sheets and are subject to change by the manufacturer. When a job depends on exact fit, ask us for a sample before the full run — we'd rather send one shirt than reprint two hundred.