That's insane. In offshore heavy lift we stick to barge motion limits around 2 degrees. Maybe we're just far more risk averse, but I cannot fathom how this would work with 8-9 degrees rotation. Even being on a ship with 2-3 degrees rotation you really feel like you're walking up and down a hill.

The barges are autonomous. The lack of people on board presumably significantly increases their tolerance for rough seas, both in terms of risk of human life and tolerance for human factors like seasickness and ability to perform labor in a bumpy environment.

That's true but there still have to be people coming aboard to seafasten the rocket. To my knowledge it's not welding itself to the deck.

If not already, this seems like a job that could be done by specially designed robots.

I agree it's insane but looking at the few previous attempts it seems its a regular expected sea state so I suppose there is no option but to work around it.

To clarify I'm talking about a double significant amplitude limit, which for a given typical three hour sea state is the mean of the highest third of double amplitudes. Does anyone know the reasonably precise location of the landing? I can then find what the typical waves are in the region.

After watching the video I don't know what Musk is talking about. Maybe he's talking about a maximum double amplitude.

Some fan calculations suggest it could handle even more, but SpaceX knows more about their coefficients of friction and center of gravity than we do.

He further mentions that they weld the rocket to the ship's deck before heading to shore.

When landings are routine, SpaceX could apply for an on land license. Blue Origin already has that.

SpaceX already landed one on land, December 22, 2015. And received a license to do so.