I've never had a question closed, but do have to agree with JPGalt that there is a problem. A lot of power is vested in high-rep users who got to high rep by making lots of snap judgments. In the Qt tag, I frequently encountered high rep C++ developers guessing at answers. They knew the language, but were unfamiliar with the library and would make guesses at how things work. Then other C++ developers browsing the front page would upvote their answer because it sounded right (even if it was completely incorrect).
It's mildly annoying when wrong answers get voted up, but it's even more annoying when moderation tools are used that way. This new user got screwed over by high-ranking close voters who didn't actually read the question: https://stackoverflow.com/q/49847677/331041
That's an interesting point. I mostly answer in Python, HTML, and CSS, and haven't seen this happen over there.
It reminds me of when SO added the feature to make front-end code snippets in answers runnable. Perhaps an evolution of that is to provide the equivalent for backend code. Do you think this would solve the voting problem you described?
To the linked question - Do you know if it's possible to undo a "mark as dupe"?
It's mildly annoying when wrong answers get voted up, but it's even more annoying when moderation tools are used that way. This new user got screwed over by high-ranking close voters who didn't actually read the question: https://stackoverflow.com/q/49847677/331041