Either the exploration makes a real difference or it doesn’t, and it’s very hard to make a real difference while also not randomising the winner. Fine for casual games, not so great for 4x marathons.
This is only a half-baked thought I came up with on the spur of the moment. It seems apparent that deck digging is not that distinct from tile-laying, so maybe the problem for competition is merely one of familiarity - it’s much easier to become familiar with a deck of distinct cards than it is a stack/pile of visually similar un-named tiles.
I think Oath has an out or two: winning may not be your only goal, and the sites and deck cycle only slowly between games, making familiarity possible. The total number of sites and the impact they have is also quite limited, while also being memorably distinct.
Eclipse, to bring this back on track, has a bunch of very different tiles that shape the game quite a lot while also not being memorably distinct, so it’s hard to get a handle on the distribution and odds involved.
Or maybe it’s not about familiarity at all - just action economy. In Eclipse, every action is important, and “wasting” one on a bad explore feels bad.