On Stack Exchange sites, being able to edit a post requires a certain amount of reputation points (depending on the site and the kind of post). Users below the reputation threshold can only submit edit proposals for review; the review process discourages making trivial edits, and there is generally a strong culture of preserving authorial intent when editing. Also, the reputation system incentivises writing a new answer rather than editing. It’s hardly bulletproof, but nevertheless this does a reasonably good job of ensuring edits, when made at all, are usually high-quality rather than low. There is rarely a good reason to revert an edit on Stack Exchange.
So if reversions are made, they are more likely to be made for bad reasons, especially by inexperienced users. In particular, the Cleanup badge is more likely to be awarded to, shall we say, conceited users starting their first ever revert war. Stack Exchange is no stranger to perverse incentives, but this particular mechanic seems especially prone to those.
(This was raised on Retrocomputing Meta, but isn’t particularly specific to that site.)