The gist of the academic studies is that, as I said, it depends. It depends on whether the item has a well known "market value", i.e. is it a new item that participants can easily determine the normal price of or a very rare item. It depends on supply vs. demand. It depends on the other bidders, are they "rational" or "irrational"? The studies suggest that sniping is sometimes an optimal strategy and sometimes it isn't.