Flipside: drastically changing a brand's sizing standards has repeatedly driven me away from long-time favourites.

This happened to me several times from the 1990s through the aughts. Literally between one shopping session and the next, the same style of clothes (tops, bottoms) which had fit perfectly no longer did, resulting both in a set of returns (of clothing) and non-returns (of myself, for future purchases) to those stores. As someone who generally dislikes the shopping experience, additional and insurmountable frictions such as these are absolutely fatal.

More recently (as I've just commented) it's the widespread adoption of stretch fabrics in non-athletic wear. I may want stretch in some of my workout clothes. I don't want it in my street clothing.

Yes. This is specifically a driver for having brand-specific sizing: knowing what size I am in Wooland Jade does nothing whatsoever to help me assess a potentially cheaper option in Uniqlo Whatever. It's the same lock-in effect as cloud APIs, only implemented through attributes instead. Imagine the chaos in the guitar market if the "bass" in "bass guitar" had up to +/-25% variation between guitar manufacturers — it would be a total nightmare trying to cross-shop guitars away from your current one, and lots of people would just end up glued to a brand so they don't have to do the hard work of assessing 'is this within +/-5% of the bass that fits me now'.

Speaking for myself as a bloke, yes, 100%. If I know stuff from a particular brand fits me well, I’m going to buy more from that brand.

Yes but it’s multiple dimensions other than just waistline. e.g Some brands make boxier shirts and others use longer cuts.

Because “my style” prefers one over the other, I know when I buy from a certain brand so it’s going to fit on me better.

If if waistlines were standardized it wouldn’t really account for all the other measurements.