Docx, XLSX and RTF are well-defined formats. And of course you can extract text from XML as well as from RTF, but you’ll lose all formatting, aka semantical information.
I was talking about the easiness with which you can work with formats. Try to change all level 2 headings to level 3 headings with a script in Docx, RTF and MD. For MD, it’s a one-liner. Or building a table of contents from level 1, 2 and 3 headings in those formats. Or find all images in a document, copy them to a sub-folder of the current one and adjust the image references in the document (we’ve been there with MD).
If one doesn’t need this kind of abilities – good. People who do, however, should look out for a format that makes these kinds of things as easy as possible. I’m not saying that MD is the silver bullet. But since it separates content from representation, it has certain advantages over RTF and XML-based formats like Docx, Pages, and Open/LibreOffice. And disadvantage, too. Of course.