Some of the accident rate can be likened to pulling your car into your garage, but it's actually not your garage at all... it's a bit different, someone changed the length of the drive, rearranged the gardening equipment, unscrewed the light bulb, and changed the intervals between those little pink flamingos or garden gnomes you use to gauge your speed into the garage. But until close to the garage itself... you still think it IS your garage. And probably some other colorful metaphors (or are those similes... no matter...)
I've no intention of taking a position on whether night flying in the offshore environment should or shouldn't happen (and I've done enough of it to have at least a passing acquaintance with most forms of visual illusions), but every risk management model I'm familiar with includes 'avoid' as a possible method of mitigating risk. It should at the least be considered, so while folks are pushing forward to develop capabilities to fly around the clock (which development is probably positive) it will indeed, at times, be in everybody's best interest to stay in bed. To spring load to 'accept' as a risk mitigation method is to lose the big picture, methinks.
There needs to be a balanced view between training, technology, administrative controls, risk management, and commercial pressure, or it's going to end badly.