And conversely if it doesn't appear somewhere that doesn't mean it wasn't part of the original. Which is why it may be better t base a judgment as to its belonging to the original on the text itself?
Nope - faulty thinking.
We have early copies of Mark (the earliest that have the ending) where the longer ending is missing. Note missing, not just that we don't have that section in the copy. The gospel ends and ends at 16:8. So we can be confident that at the very least some early versions of Mark didn't include 16:9-20. Conversely we have no early copies of Mark that include 16:9-20. Could be, of course, that we haven't found an early copy with 16:9-20, but that these exist or existed.
So at best you may be able to argue that early on there were variants circulating, one without 16:9-20 (we know this for sure) and one with 16:9-20 (we cannot be sure about this. So if you argue this, then you'd have to try to discern which was more likely to be the original. So if the one including 16:9-20 was the original then you'd have to argue that it was removed from some copies. But this is a pretty hard ask because:
1. We don't even know for sure that any early versions of Mark included 16:9-20, but we know that some didn't.
2. It is easier to argue for 16:9-20 to have been added, rather than removed on the basis that its addition adds theoretical kudos to the gospel, while removing it would weaken claims.
3. As Jeremy points out there is also linguistic evidence that is isn't original.
So taken together we have strong evidence that 16:9-20 are later additions and not in the original. The evidence that they were in the original (which would mean they were somehow removed from some early versions) is very weak.