I am struggling to understand why linked lists seem to be accepted to have better big-O performance than arrays for in-middle insertion/deletion. The performance differences between lists and arrays for front-of and end-of operations make sense intuitively, but for operations in the middle, not so much.
I see why the middle insert is O(n) for a dynamic array - half the elements need to be shifted one by one.
But, when it comes to linked lists, don't half the elements need to be traversed one by one starting from the head to find the insertion point? And if so, why doesn't that also make linked lists O(n) for middle insertion? And if it does, why is it not represented that way in the Wikipedia article?