It may well be one of the best preserved monuments of Ancient Greece, but the projects to protect the Temple of Apollo Epicurius both help and hinder your ability to appreciate it. Covered by some kind of tent since 1987, you unfortunately can't get a wider vista of the temple, and you also can't see how it fits in with the broader landscapes up here in the mountains (which is particularly disappointing because that was presumably a big factor in why it was built in such a remote location).

However, I found that being forced to stand close to the temple created a sense of intimacy that I haven't found at other ancient sites in Greece. With the tent around us, I felt even closer to the monument and it seemed larger than it might have otherwise, the columns looming above. Rather than admiring its overall scale from afar, I had more incentive to study the little details that might otherwise have gone unnoticed.

Much of the OUV is in these details and, although there are some information signs, I would suggest learning a bit about them in advance or bringing something to read with you. Luckily I had an archaeologist guide who was able to point out the important design features and talk about some of the theories about why it was built north-south (rather than the usual east-west) for instance, or about why there was a prominent Corinthian column in the centre (maybe there there was actually no statue of Apollo and this was its proxy?).

It's not easy to visit the Temple of Apollo Epicurius without a car and I would probably recommend driving rather then trying to do this with public transport (I did it as part of a tour with a minibus). If you were to do public transport, the suggestion of basing yourself overnight in Andritsena and doing the hike to the temple does look quite nice, but I can't comment on it from personal experience. But having a car is much more efficient and means you can also combine a visit to the temple with some of the other nearby World Heritage Sites in the Peloponnese like Olympia and Mystras.