All brands

Fiber labels on 44 men’s brands: what our dataset shows

FabricTruth keeps a table of the fiber content of 120 core styles from 44 men’s clothing brands: tees, shirts, base layers, underwear and socks. On November 2, 2026 we re-read every one of those labels on the brand’s own product page. The 89 labels we could read, from 37 brands, are what the numbers below count. The other 31 are flagged in the table and left out. Every number is computed from the table at build time, and the full table is published as a CSV.

These describe the 89 checked styles from brands we chose; they are not a random sample of the market. Read the limitations before quoting them.

What we could verify

We opened each product page on November 2, 2026 and copied the fiber line as printed. Of the 120 rows:

• 67 matched the page.
• 11 didn’t match, so we changed them to what the page says.
• 11 named a product we couldn’t find on the brand’s site. The brand sells a product of the same type and main fiber, so we read that one instead.
• 31 couldn’t be verified: 9 page blocked our reader, 4 page names fibers but gives no percentages, 1 page lists no fiber content, 15 product not found on the brand’s site and 2 store closed or product gone. They are marked “not verified” in the table with no figure, and don’t count in any number on this page.

In all, the reading changed the natural-fiber share on 15 rows. Each change is shown in the full table below and in the CSV.

By category

Share of checked styles in each category whose label is 100% natural fiber. The table adds the average and median natural share.

Across the 89 checked styles: 50 are 100% natural, 27 (30%) are part natural and part non-natural, and 12 (13%) have no natural fiber. Nylon or polyamide appears on 18 styles and elastane, spandex or Lycra on 31.

How the 26 scored brands score

Scores don’t come from the table above. Since November 2, 2026, a brand’s score is the average natural-fiber share of 6 to 8 styles picked by a fixed rule from the brand’s own best-sellers list, 218 styles in all, and a brand needs at least 5 of them across 3 garment categories to get one (sampling rule and formula, score sample CSV). 18 brands are unscored for now: Darn Tough, Asket, Pact, Pendleton, 3sixteen, Left Field NYC, Lands' End, Flint and Tinder, J.Crew, Uniqlo, Cotopaxi, Outlier, Western Rise, Bluffworks, Vuori, Lululemon, Under Armour, Wayve. Number of scored brands in each band:

Method

• The dataset is the composition table behind every brand page:1 to 8 core men’s styles per brand (120 in all), each with the fiber content from its public composition label and a natural-fiber percentage.
• On November 2, 2026 we fetched each row’s product page and recorded the fiber percentages as printed, the page address and the date. Where a page was blocked, gave no percentages, or the product was gone, we marked the row “not verified” rather than guess. Most pages were the brand’s US store. One reading (lululemon’s Always In Motion boxer) came from its German store. Two Tommy John readings came from the store’s own product data for the linked page, because the page itself lists fibers without percentages.
• Natural means cotton, wool (including merino), linen, hemp, alpaca, cashmere and silk. Polyester, nylon, acrylic, elastane and similar synthetics are not natural, and neither is regenerated cellulose (viscose, modal, lyocell, “bamboo” rayon), which US rules define as a manufactured fiber².
• Percentages are by weight, as US labels state them, listed in order of predominance¹.
• We counted rows in code from the site’s data file. “Polyester appears” means the label text names polyester; it doesn’t mean the style is mostly polyester.
• Download the full table: fabrictruth-label-dataset.csv (120 rows). Columns give each row’s check status, whether it counts in the numbers here, and the product we first listed where the reading replaced it.

Limitations and counterpoints

• Not a random sample. We picked the brands, and the styles are a handful of core pieces per brand, not full catalogs. The site was built to cover natural-fiber brands, so they likely make up more of our list than of the market.
• Not every label could be read. 31 of 120 rows are left out, and 7 brands have no checked row at all, so those brands don’t appear in these numbers. Several are large mainstream retailers whose sites blocked our reader, which may tilt the checked set toward smaller brands.
• Replacements change the sample. 11 rows now describe a different product from the one we first listed, because the original wasn’t on the brand’s site. We picked the same garment type and main fiber, but it is still our choice.
• Labels change. Brands reformulate and rename products. A label is only what the page said on the date we read it.
• Labels can understate small fibers. Under FTC rules, fibers under 5% may be grouped as “other fiber,” trim such as waistbands and cuffs is excluded, and elastic under 20% of the surface area can be left off with “exclusive of elastic”¹. So some “100%” labels may carry small non-natural parts.
• Labels have a tolerance. The FTC guide allows a 3 percentage point tolerance on fiber content claims ¹, and we didn’t test any garment.
• Natural isn’t automatically better. In one wardrobe survey, synthetic garments were reported to last longer (6.3 years) than 100% wool (5.3) or 100% cotton (3.6). That survey was commissioned by the wool industry³. See our blends comparison for what the synthetic share does well.
• Scores rest on a sample. A score averages 6 to 8 best-selling styles per brand, not its whole catalog, and a best-sellers list is what the brand chooses to show. 5 brands score 100 because every sampled style was 100% natural; that doesn’t mean everything they sell is. Scores are not lab assays, certifications or ratings.

Sources

1. 1US Federal Trade Commission. Threading your way through the labeling requirements under the Textile and Wool Acts. Business guidance (archived edition on govinfo.gov; the FTC’s current web copy blocked our fetcher), c. 2006.
2. 2US Federal Trade Commission. Generic names and definitions for manufactured fibers (16 CFR 303.7). Code of Federal Regulations, Textile Rules, current.
3. 3Laitala, Klepp (SIFO, OsloMet), Henry (QUT). Does use matter? Comparison of environmental impacts of clothing based on fiber type. Sustainability, 2018. Co-funded by Australian Wool Innovation and the Research Council of Norway; lifespan data from an AWI-commissioned survey.

Published November 2, 2026 by FabricTruth (run by Rexmark LLC). Labels re-read November 2, 2026. Numbers update when our brand data changes. You’re welcome to cite or reuse the numbers and CSV with a link to this page. Corrections: [email protected] (see our corrections policy).