It's pretty well known that tees made using 1920s-era loopwheeling machines don't suffer from size changes, and age really well. But sadly these are now super expensive, and only offered by Japanese niche brands who bought machinery from American corps.
I have been thinking the same for a long time. Clothing these days are just branding, marketing as such. I don't want any of that. I want quality, decent materials.
Like the article suggested, materials in clothing these days cost very little. I used to own two Ralf Lauren Polo Cotton T-Shirt. One Cost $200 the other $100, both were a sensation to wear. I still have them today and i love them. The problem is i got them as gift. I could never afford to buy them myself. I just wanted simple styling and plain colour, nothing fancy, and decent quality materials like those Polo Shirt i have. I tried a few Start up who claim they uses premium quality but they surely dont match those.
For now i continue to stick to Uniqlo for pretty much everything. They sure are not the best, but I do find them value for money, at least comparative speaking.
They're £58.33 plus £10.00 shipping which converts to roughly $100 US. I'm not that frugal, but Jesus this shirt better be tip top quality and last longer than 6 months for $100.
The names are used in Japan. Some of these names (e.g., "Strike Gold") conjure up an old-school Americana. This is intentional. Japanese denim brands (who now make these shirts as well) are very reverent of pre-1960s Levis.
They use the English names in Japan. If you google for the Japanese official websites for these brands or even their products on rakuten.co.jp you'll see the brand names in English.
I remember visiting a Japanese clothes shop here in Dubai and I noticed their shirts were quite different in texture than anything else I'd seen. It was expensive too. I think it's called Superdry? Could that be one of the niche brands you're talking about or are they just using a unique blend of textiles?
In any case, it's quite fascinating how sometimes old tech is simply better than what we have today.
I have lived in SE Asia for half a decade and am currently in Taipei and honestly I see a lot of 15-30 year old Asians wearing Superdry. They invariably think it is actually Japanese. I haven't noticed older guys wearing it here but perhaps I haven't been looking.
Superdry clothes have a reasonable price/quality ratio. I'd rate the clothes higher quality than any other Jack&Jones/Tom Tailor/H&M etc brands I own though.
It's quite funny actually: "The Japanese text incorporated in the brand's logo—極度乾燥(しなさい) (kyokudo kansō (shinasai)?)—literally translates as "Maximum dry (please do)", the text in brackets being due to the translation software used offering alternatives depending on whether dry is intended as a noun (e.g., super dryness) or an imperative, (e.g., dry this shirt out)." [0]
I fins the fits really strange, especially for men. Everything there is a tall fit. And the quality is rubbish, I had 2 of the zippers on my winter coat to break after a couple months use. I expected more for a decently priced coat.
> While hot water may cause shrinkage in wool garments, for cotton and polyester t-shirts, the washer settings don't make a big difference.
This is also more or less true of wool, in fact. It's not the hot water that shrinks a wool sweater, it's getting it damp and agitating it. The hotter it is while agitating, the faster it will shrink; being fully wet (as opposed to damp) mitigates the felting process somewhat. It appears that cotton is the same way (although its shrinkage is less extreme and less permanent than shrinking a wool sweater!). You can certainly get a wool sweater wet with hot water without shrinking it at all. If you're careful, you can wash it in hot water and agitate it to wash it properly, as long as you don't overdo it, without shrinking it. With some treated wools you can tumble dry on no heat, although I wouldn't recommend it; but under no circumstances should you wash a wool garment (in whatever heat of water) and then put it in a dryer with any heat at all. That will shrink it.
I knit, so I have extra awareness of how many people just don't know what to do with wool these days, and I have to educate them if I want to give them something I made. :)
I'm the founder of threadbase. Thanks everyone for your kind words. I'd love to hear any comments or suggestions for what you'd like to see next or how we can improve the user experience. We're also looking for front-end/design help, as well as help with computer vision tech. Feel free to email me chris@threadbase.com.
I understand the desire for adults to have a shirt that fits, but I think that falls squarely in the bucket of "nice to have". As a parent of a child that doesn't fit any of the sizing charts, being able to predict what size of what brand will fit my kid would be a "must have"! Kids also tend to go through clothes a lot faster than adults, making the potential savings of getting a good fit far greater.
Is it as important that kids clothing fits precisely though? I mean, evidently it is to you but I would I would have thought parents would be happy that the kids look decent enough. You even point out their rapid growth so it's not like they're going to be continuously wearing the perfect fit all the time. Buy slightly big and they grow in to it.
It's not so much "precisely" as "not ridiculously wrong". The difference between "slightly big" and "tent" isn't much, considering there is practically no objective basis to kids' sizings, which are basically the intended age with no consideration of variations in age & growth; the sizes do not reflect an actual size.
Yes, this! Just to get close to fitting, we have to get our son pants for a 6yo, shirts for a 4yo, and socks for a 3yo, and even then we end up rolling the sleeves on his shirts because he needs the length in the chest but not as much the arms. I can't tell you how many pants, shirts, socks, etc. we went through before figuring out this is the combination that works. Oh, yeah, and this is only with the specific combination of pant/shirt/sock brands that he's currently wearing. If we had to replace one part with a different brand (which we will, because he's growing and for some reason there's no brand that reliably stocks the full toddler-kid-youth size range), we'd be screwed.
Most important (for us, anyway) is t-shirt quality and softness. Like you said, we buy big and they grow into it and out of it but it's the quality that really matters. We buy mostly Hanna Andersson shirts for our boys. They're expensive but they still look good after two seasons of wear, which is remarkable for kids' clothing. Our younger boy wears many of the HA shirts and coats that his big brother once wore.
This applies more to shoes than clothes. It's ok to have a little bit of variance in clothing, but given how much kids run around, they need shoes/sandals that fit really well.
I organize an annual conference [1]. All of our attendees get a free tshirt. We purchase all of our tees from a single manufacturer (say American Apparel) and let the attendee tell us the size they want before we finalize the order.
The problem is that attendees often don't know what size will fit them, and sometimes guess wrong and end up with a poor fit. Anything you could offer us to help here?
They should build a system like the "True Fit" size finder on Moosejaw.com. You tell them about your body size and then give them some brand/size combos that you already have and like. The finder can then tell you which size to buy for XYZ brand.
Excellent work and presentation! I'd love to see a similar treatment on jeans. For instance, I wear Levi's, and always default to 501 32/32, and the 5 pairs I have range from baggy to painted-on. I'd be interested to see the manufacturing variance within/between countries of origin–I have pairs from 3 different countries, and like I said, very different fits.
It would interesting to see the effects of washing on denim jeans since there was a huge uproar when some famous designers came out and said the NEVER wash their denim jeans since it ruins the material:
This is great for comparing cotton or cotton blend shirts. I'd love to see the same data for shirts that don't use any cotton. We make fabrics for our own brand and usually stay away from cotton partly due to the problem of fit variance after wash.
Threadbase is a really cool project. I'd definitely use something like this if it were available in my country. Good luck, I hope you succeed and go global.
Then you finally find a shirt which is just the "right" length at the store, and three washes later and it is too short... But you cannot buy the next size up since you need more LENGTH not shoulder width.
Very few brands make t-shirts which are 2x-tall (as opposed to 2x-large, where both the length AND width increases). A medium-2xtall would be like the unicorn of t-shirts.
The sizes should be: S, M, L, 2x, 3x, etc & S-T, S-2xT, S-3xT, M-T, M-2xT, etc.
There's a Dutch brand that specializes in extra long t-shirts: http://www.girav.com/ - they have Standard Long for men 1.75m (5.8 ft) to 1.85m (6 ft), Long Fit 1.85m (6 ft) / 1.90m (6.2 ft) or taller, and
Extra Long Fit 2.00m (6.6 ft) and over.
I'm a huge fan of Eddie Bauer's medium-tall shirts. They usually don't sell them in store, but you can get one online if you are looking for an option to try out.
It took me forever to find a fairly athletic fit (slim but not skin tight) in my height (6'2" with a long torso) that also doesn't have logos or random graphics (solid colors and maybe a pocket is all I want).
If you have a favorite choice it'd be great to know it and have more options.
*Full disclosure: I worked at an Eddie Bauer in college.
This is how military utility uniforms are sized: Small, Medium, Large and Short, Regular, Tall. I'm surprised that idea hasn't spread into mainstream sizing.
I feel obliged to share a lifehack my mother taught me regarding the laundry.
After removing the laundry, take your t-shirts and stretch them yourself, one by one, when they're still slightly wet. Grab them with two hands symmetrically, stretch horizontally, moving your hands down along the shirt. Do the same vertically, and with the sleeves.
Do not use a machine dryer, just a regular standing dryer like [1].
Put your t-shirts carefully, symmetrically on the dryer, and once dry, put them on a hanger. If you follow this, you will not have to iron them at all.
Source: been doing this for 4 years and I didn't touch the iron since. I have all t-shirts 100% cotton (though I buy only high-grammage ones) and they all seem brand new and ironed (the only exception being one particular brand whose collar looks bad unless ironed, I stopped buying that brand). YMMV of course.
Just before hanging them, while they are still wet:
Grab each t-shirt from the top with both hands, then do a couple of violent whip-like movements. This has the same effect of stretching the fabric so that no ironing is needed.
If you do it right, you can get a nice crack sound. I've found that I can usually get at one from a freshly washed item of clothing (and towels) and it only cracks once per fresh wash. I don't understand the mechanism that causes it but find it satisfying for some reason.
Only if you have a rubbish washing machine that spits out very wet clothes. Mine is a front loading ASKO which spins the clothes to the point of being touch dry. I give each t-shirt a whip-shake and dry them in a cool tumble dryer.
> I feel obliged to share a lifehack my mother taught me regarding the laundry. After removing the laundry, take your t-shirts and stretch them yourself, one by one, when they're still slightly wet. Grab them with two hands symmetrically, stretch horizontally, moving your hands down along the shirt. Do the same vertically, and with the sleeves.
Interesting, I never considered this a hack. Noone told me this, but I always do it. Ironing is never needed. I do it with pants too.
On a less sarcastic note: Water has no problem evaporating at room temperature. A batch of clothes is usually dry overnight. In fact you can even dry stuff at below-zero temperatures through sublimation.
The trick I was taught was to dry the light clothes (shirts, underwear, socks) separately from the heavy clothes (jeans, hoodies). Then hang the shirts up immediately after pulling them out of the dryer while they are still hot. Then you won't need to iron your shirts. It may not work really well for some dress shirts though.
My wife has a similar trick that we've been using successfully.
After removing the laundry, we fold everything neatly, and leave them stacked on top of each other for 30-60 mins. Than we dry them with a standing dryer.
Haven't ironed t-shirts or jeans since we got married.
I've been using the same technique for a while and it's working quite well. Unfortunately, it's only for tee-shirts and not shirts, so the iron is still required from times to times.
Great to have this data. But wouldn't the extent of dryer-induced shrinkage be driven by the amount of time in the dryer, as much or more than temperature?
While most modern dryers offer a choice of temperatures, the big knob mostly controls a humidistat-based target. I personally equate the "very dry" setting with "shrink beyond usability".
I'd expect that removing clothes while still damp would be more important to avoiding shrinkage than reducing the heat, but I'm no T-shirt scientist. (T-shirtician? T-shirtologist?)
"The biggest determinant of shrinkage is whether the shirt went in the dryer or not. (We wash and dry all t-shirts using a warm wash and normal/warm dry cycle)."
That doesn't answer the question, which is an interesting one. Where in the dry cycle does shrinkage occur? If you dry on hot to slightly damp, do you get the same level of shrinkage as cool to fully dry? If so, why haven't I ever seen a dryer that starts hot and transitions to cool as the clothes dry out?
If so, why haven't I ever seen a dryer that starts hot and transitions to cool as the clothes dry out?
I'm pretty sure that all of the three dryers we've had in the last twenty years had a permanent press setting that does exactly that. Our current one certainly does, and it's just the front-loader that was on sale at Home Depot, nothing fancy.
But they call it "permanent press" so those of us who didn't study domestic science don't know what it does ;)
Yeah, fair enough. Come to think of it, I don't even know why I know this. Our current dryer has something on the display that says something to that affect, but our previous ones were old school analog dials without explanatory text, yet I always knew that "permanent press" meant it starts out hot and ends cool so the clothes don't wrinkle.
Good stuff. I'm interested to see data on other cotton garments, particularly buttondown shirts.
Cotton is just a lousy fiber. On the other hand, wool is a strong and resilient fiber. It also never needs to be washed provided it isn't stained.
My wife knit a wool sweater for a close friend of mine who spent 6 months as a bosun on the tallship the Lady Washington (the Interceptor in the Pirates of the Caribbean). Fresh water is scarce on a tallship so showers were infrequent. He came home during Christmas and I smelled the sweater which he claims he never washed and it smelled fresh. Surprisingly, it also kept him warm and dry on the open ocean. I later learned that Irish fishermen have been wearing wool sweaters at sea for generations.
"Cotton is just a lousy fiber. On the other hand, wool is a strong and resilient fiber. It also never needs to be washed provided it isn't stained."
You're mixing things up. Cotton is stronger than wool, cotton shirts last many years longer than wool shirts. It's less noticeable with sweaters simply because their fabric is thicker.
On the other hand, wool can be more comfortable to wear because it retains more heat and absorbs moisture better.
The "manufacturing variance" chart jumped out at me as looking fairly unnatural: there's a variation in width or variation in length but very little points that mix. Then I noticed that we're talking about just over half an inch in each direction.
How much of this effect is variation in your measurement?
There's a standard process we have for measuring each item. The goal is to flatten the shirt without pulling on the material.
Measuring clothing is really, really hard to do in a consistent way. In my original plan for the project, I was going to hire 5 task rabbits to measure everything, then throw out the min and max and take the median of the remaining three numbers. It didn't work. The average taskrabbit had a standard deviation of error of 0.3" for each shirt. That might have been OK if the errors were purely random, but there was a lot of systematic error as well.
Eventually I was able to find two people who could provide consistent measurements. They were within 0.1" of my measurement for each item. When there was a difference of more than 0.1" we would remeasure the shirt. In general I feel good about the data we published.
This is excellent. Independent, consumer-empowering size analysis, especially measured over extended wear and washing. I'll be following you.
I wonder if there are any companies that sell inexpensive custom t-shirts? Provide your measurements, specify desired fit, neck type, color, and fabric, and order exactly what you want.
As someone very hard to fit for pants (28" waist and cyclist thighs), I would be thrilled if you also do this for jeans and shorts.
We've thought about doing made-to-order tees before, but at the end of the day, since we sew in Los Angeles, they'd end up being like $75 each for a custom pattern.
There are some companies that have done tees and polos though - Polos - VASTRM. Button ups - Blank Label, Pacific Issue, Proper Cloth. Tees - SonofaTailor, Threadmason.
I've standardized my shirts on the Uniqlo Crew Shirts and they seem to have very little change from wash to wash compared to my Banana Shirts.
The Uniqlo shirts are a cotton/polyester blend while the BR shirts are 100% cotton, which explains the help that durability of the synthetic materials provide.
It would be great if we could get some data on which brands have the most and least variance and which brands expand and shrink the most over their lifetime.
Poly blends are not going to noticeably get smaller wash to wash. They will get worse-looking though, as they don't shrink consistency - only the cotton shrinks, not the polyester.
I don't have a dryer, but I've still had cotton shirts (not t-shirts) shrink in the washing machine. This usually happens the first time (or first few times) they are washed at the temperature recommended on the label: 40°C (104°F). However, at 30°C (86°F) I've never encountered any shrinkage. So this purely anecdotal experience makes me believe that the temperature of water can affect some cotton garments.
I've stopped buying 100% cotton t-shirts because they shrink so much. polyester/cotton or tri-blend last so much longer. Also, I find them more comfortable to use (cotton sticks to the skin when it gets humid).
FWIW you personal website linked from your profile doesn't either:
Fatal error: Namespace declaration statement has to be the very first statement in the script in /home/thebarproject/webapps/codecon_wp/wp-content/plugins/zencache/zencache.inc.php on line 13
To be fair, that's a server-side error that will probably always occur (so the site is effectively permanently down), not like the original client-side error caused by users configuring their browsers in a way that site owners might consider hostile or otherwise not worth supporting.
It's not considered hostile by the site owners https://news.ycombinator.com/item?id=10959363
in fact it's already fixed. My original comment was in the spirit of helping a fellow programmer find a possible bug. On the other hand I found themartorana's comment to be particularly snarky wrt a large proportion of web citizens that are increasingly worried about their privacy.
The fact that his own website doesn't seem to work at all is just very ironic.
I do see that these particular site owners don't mind their users using Ghostery and uBlock. Some may not go out of their way because they see it as their users denying them ad revenue or analytics (the user is "hostile".) A similar situation happens with NoScript, where some site owners view Javascript as a basic, easily-met prerequisite to using their site, and will do nothing to help users who don't want to enable JS.
I presumed that's what themartorana was referring to about why they might not bother to fix it, and why I thought it would be useful to point out that according to the error message you got, themartorana's site is broken in a different, more fundamental way.
This is a cool analysis, although I think every t-shirt I own is from Target (Merona and Mossimo brands) so I didn't have a single point of reference for the width and length charts. Those charts seemed like by far the most useful part of this post.
Edit: I'm curious about the downvotes. Are people appalled at my lack of taste in t-shirts? :)
The fabric can't always be rotated 90 degrees. Some T-shirts body portion is made from tubes of fabric. The ones with seams on the sides have the possiblity, but that highly depends on the weave of the t-shirt material itself. The material is knit in a way that is stretchy, and turning the pattern pieces (the only logical way to turn the fabric 90 degrees) can make the shirt not stretch the right directions when worn and/or force changes in the garment itself.
I'm not sure that it would.. the next sentence says that the shirt expands when worn and 2x more in the chest than the length.. that says to me that its not the weave, its just being stretched more in the chest direction so the length never recovers from its shrinking which becomes cumulative.
I think this is because the shirt is stretched horizontally just from being worn but obviously there's no forces stretching the shirt vertically because any force that would stretch it vertically just raises the bottom instead
I own an online apparel company that I launched on HN three years ago, but before this, I built tech companies, so maybe I can weigh in with what I've found (and have been really surprised by).
Fit is almost a philosophical decision. Most companies only have five sizes (XS-XL) and the goal is probably to accommodate as large of a % of your customer base as you can with each of those five sizes. Most designers either use themselves to build their sizing from, or use a fit model, someone that literally is paid to stay the same size year round and give input on the fit.
After the decision on who, exactly, you want to design to, you need to decide how the individual garment should fit. eg. some streetwear might have a 4" longer shirt from the high point shoulder than mine, if they like that long, baggy look. Some styles, like James Perse, may want a laid back, beach vibe, or cater to an older demo that might want a more forgiving shirt, so they'll want a few extra inches around the waist.
In our case, I'm a very consistent medium -- almost every brand I'd expect my friends to buy from fits me well at a medium -- so I tailored our shirts to myself. We then started with what is a fairly standard grade rule to go up and down the sizes, and then after many, many batches, fine tuned each size to feedback from our customers.
So -- a lot going on. There is no way to standardize what a "medium" is across brands (and some companies don't even use the same terminology!)
The frustrating part, as a consumer, however, is when a brand doesn't have consistency between batches of the same style, colors within the same style, or between styles!! eg. I have a pair of Levis (huge fan..) on right now that I have the same exact wash, and style, but the 31 waist is MUCH bigger than the 32 waist. In this case, they were probably sewn in different factories, and they don't have tight enough tolerances in QC, or something. Between styles happen because a big company like Nike, for instance, may have 100 designers, who may not want to re-use design elements/blocks from one style to the next - and would prefer to fit / design their new tee how they see fit. But as a consumer, I expect all Nike medium shirts to fit the same, so that is frustrating.
Turns out, sewing things by hand, and dying, pre-shrinking natural fabrics, etc. is really hard. We've been working on our processes for three years, and are incredibly proud of what we've built... but we learn new ways to improve every day!
I would love to see the size dataset expand to casual and dress buttonups, and even jeans. A bit of data like this will greatly improve my shopping experience.
The most maddening things is that even within a brand, size, S,M,L, etc. are inconsistent, never mind hoping there would be consistency of measurement across brands.
Their charts expose this onconsistency. Some brands Mack Weldon, is more consistent than say American Apparel.
Even when objective measures like pants waist size in inches, typically a size 32" actually 34" --I guess to make people think they are thinner tan they actually are.
Wash as usual and then hang on a plastic hanger until dry. So, don't use a dryer and just let them hang, starting when they are still wet from the washing.
Also works when hand wash and rinse but don't squeeze out much of the water, that is, hang them while they are still wet enough to drip.
I learned the dryer effect hard way - so many good tees were destroyed. :( Nowadays I just leave my t-shirts and polos to dry.
The size charts is a real eye opener. I know that Abercrombie carries smalls that fit me fine, but Zara was an unknown to me. Apparently, their tees are also reasonably priced and look pretty good...
Interesting data! I remember the excessively short/wide t-shirts well from my early university days, when I still took my laundry to my parents who had a dryer. Now I dry my clothes on a line, and they don't seem to deform so much.
The conclusions are in the headlines and further details can be found in the paragraphs below. I think it was easy to skim through and get a good picture of their takeaways.
Is there a "grain" to the fabric or something? Why not turn it 90deg and have the shirts increase in length and decrease in the chest instead? I'd prefer size to stay the same over time, but if I had to choose I think I'd rather have that.
I still have a problem with my big and tall shirts - I have to hang-dry them to keep them from shrinking vertically when run through the dryer (due to my body type). This backs up what I've complained about for years :)
The most important thing I learned from this page was that American Apparel sizes small. Since 3/4s of all my shirts are startup shirts and most of those are AA, this is good to know.
The knit pattern (tees are not woven) only affects the ratio of stretch per direction. Shrinkage is much more a property of the fiber and fabric processing, like prewashing.
