Really? I tried twice playing with epic speed. The first time I had a perfect start with Egypt, starting next to an atoll tile and getting culture from ruins to get the progress opener. Right after that found a maritime city state, growing from 1 population to 3 in three turns and getting 60 science towards the wheel. The start was so powerful that had I wanted it, I would have had 5 population before the 2 different civs I had met had 4 population in their capitals. Well, I didn't want that but instead rushed for stonehenge and would have had it on turn 29. Instead, I looked and saw how long it would take from the AI to build it. Another civ finished it on turn 35.
I had another go, this time with Korea, as I wanted to see how a civ without bonuses could fare with the AI. The start wasn't as good as ruins didn't really give me anything useful and I only got faith from city states. I would have had my stonehenge ready on turn 39, but AI beat me to it on turn 25! Even with a nearly perfect start with a perfect civ to build the stonehenge it would have been impossible for me to make it in this game. There is no way it could take 77 turns for the AI to build stonehenge, unless you happen to face only civs that don't care about religion. Even then I wouldn't find it likely.
And we come back to my earlier point: I could not base my strategy on rushing for an early wonder that I wouldn't be likely to get. If I don't play with Egypt, I'd have to accept that most likely AI would beat me to it. In the likely case of failing, I'd have to either keep on starting over, which is a very cheap or just accept losing production from several early turns, which i could have used in a meaningful way, for example building a granary.
I can agree with the argument that changing the game pace makes a difference. I remember playing a version of CPP on summer or spring and trying both marathon and standard. Marathon was so much easier. A friend of mine also tried quick and had major problems, which he didn't have on standard pace. We concluded that the faster the pace is, the harder it gets. I guess it was because a human player could take a bigger advantage from moving his/her units. I'd also say that it is easier to build stonehenge on epic than standard, but after my two tests I just don't see that I could be likely enough to build it to ever try.