Adobe Acrobat and Firefox's PDF viewers, instead of numbering pages from 1, actually use some numbering schemes based on page ranges. These ranges are embedded into the PDF, so they it makes sense use them by default.
This becomes a problem for Edge when the numbering of the pages doesn't match with their indices in the pdf. So for example, this "Sherlock Holmes" book has page 8 as the 14th page in the pdf. The PDF has the numbering ranges, it has the information to fix this, it just needs to be used.
It is important to note that when you write down the page number you want in the "Page Number" text box it should navigate to the page after de-mapping it from the numbering scheme, i.e. after finding its index given the numbering scheme. Otherwise It becomes terrible to navigate pages using Tables of Contents inside pdfs when the number on them is different from the actual page number.