Simple example: If the site uses JavaScript to produce its pages (like Apple’s developer documentation does). The webarchive will contain the JavaScript code, but not the content that is generated by it. Obviously, because that depends on the code being executed which it will only be when the page is displayed by a browser (or the integrated WebView in DT).
So the JavaScript code remains the same, e.g. it does something like load item 12345 from server xyz. Now, item 12345 might contain “Hi there” today. But tomorrow, someone changes that to “Go away”, and than that’ll be what you see.
I tried to summarize the Pros and Cons of the different HTML download options in a post here