There is a problem with the number of dimensions. Even a t-shirt is described best in neck, chest, waist, and you could add several parameters for sleeves and also for heights. Neither consumers nor manufacturers can really handle the combinatorial explosion, so you have to boil it down to one or maybe two dimensions.

But the reason even that isn't done is mostly history and market expectations. There are clothing categories that sell in actual dimensions, and (aside from the terrible dimensional accuracy of clothing in general) it works fine. But those are all on the "men's" side, and it seems the industry believes women will not like buying based on actual numbers.

Sizing accuracy would definitely be an improvement, but it would shine a spotlight on the core issue that clothing manufacturers are hiding from consumers: they only produce hourglass shaped clothing and ignore the vast majority of body types altogether. If everyone’s sizing was 48-38-48-15 (bust-waist-hip-crotch) rather than L/XL/2XL, at least it would be a lot simpler to write a search engine for it — but there has, in the past century’s history, been extreme hostility from U.S. women in finding that their bodies are changing shape as they age, and so retailers are forced to choose between a sharp drop in profits by telling the truth, or making it difficult to cross-shop between different retailers. As no regulations exist around this, they appear to uniformly choose the latter.

I would have said that this is covered under vanity sizing. I have the feeling some consumers might be more inclined to buy a product if it has size M instead of XXL, even if the product is exactly the same. So my take is that, in the name of profit, companies lie about size.

Yeah it's weird that mens pants and womens bras have numerical sizing but other womens clothing doesn't.

They can lie with those too: https://www.reddit.com/r/fatlogic/comments/8srelb/info_graph...