> The sad part is that when I've tried to remove obviously flimsy information, someone will immediately come along and add it back.

The trick is to write about your proposed edit on the talk page and wait a few days. If nobody has complained, you make the edit and write "see talk" in the edit summary. The notion that you should push an edit first and wait for someone to revert you just doesn't work in practice except for trivial typo fixes. Discuss your edit in depth, then push it once you have a presumed near-consensus for it.

I think it's important to edit early and often, but it certainly can't hurt to also explain your edits on the talk page. Bonus points if the other side makes no explanations, you get to "rv unexplained edit, see talk page". Just look in on the article every couple of days for a while to see what sticks and what doesn't. Originally when I started editing, more often than not people would have improved and built on my edits, rather than fought them. But you may need to be a bit (un)lucky these days?