Wikipedia[1] cites a 2010 study that gives a boiloff rate of 0.1% per day for hydrolox. Not sure if it would be less for pure LOX. Methane would remain stable so I wonder if there's a way you could balance the fuel load so that you can deliver only methane on some of the trips and then all the LOX in one trip to avoid having it hang around in orbit.

[1] https://en.wikipedia.org/wiki/Propellant_depot#Feasibility_o...

Even if they need to refuel quickly, why not launch 6 starships for that mission? They’re planning to build lots of them aren’t they?

As long as the crewed starship is fuelled in orbit not too long after the last tanker ship launched, then the amount of boiloff will be negligible.

I suspect Spacex will develop cooling systems to re-liquefy any gaseous propellants as this will be more of a concern for the ~6 month mission to Mars.