For me its the commodities.
I grant that SpaceX engineers are smart people and can figure out how to make Starship and Superheavy reliable and reusable.
But if they have to launch 10-14 times in order to get the propellant to the LEO depot in order to fuel the Lunar Starship, can we actually deliver that many launches worth of LOX and LNG to the launch pads in the timeframe needed to prevent it all from boiling off once in orbit before Lunar Starship can get there, get refueled and head to the moon? I don't know the answer to that, and to me that seems like the hard problem.
> But if they have to launch 10-14 times in order to get the propellant to the LEO depot in order to fuel the Lunar Starship, can we actually deliver that many launches worth of LOX and LNG to the launch pads in the timeframe needed
If only Starbase was located somewhere near abundant gas pipelines, within spitting distance of of the Texas Shale Oil boom…
When Korolyov worked on N-1 rocket in 1960-s, some plans included building a hydrogen upper stage. http://astronautix.com/n/n1blocksr.html Hydrogen is rather hard to keep cold, but that stage was designed to work for over 11 days.
Falcon-9 flies almost every other day, about 3 times per week. Methane is way more storable than hydrogen. Of course we'd like to compare numbers, but, given that Starship is way bigger than than N-1 stage - about 15 times, and there is the law of squares-cubes, which for our case says the bigger the tank the less percent of boiloff per unit of time, and it's methane, and we can afford to lose a little and top off with another tanker...
Now, how many tanker flights we'll need? That's a favorite riddle in Musk's plans :) . Korolyov, again, had some early ideas for 5 tankers - https://graphicsnickstevens.substack.com/p/sever-the-bridge-... ... For Starship - if you have 1500 tons of fuel in the Starship, and 150 tons of payload in a tanker, you need 10 flights. You can probably optimize, or be disadvantaged by some obstacles - so, 8-12 flights? That many can fly in less than a month. We can also use additional measures to reduce boiloff - better protection from the Sun, active cooling, maybe more permanent orbital refueling depot - but still, with our today's Falcon-9 flight rate we may consider one Starship per month refueled on LEO. Even if some refueling flights won't be successful, the replacements could be sent.
I personally suspect Starship will fly much more often than Falcon-9. We're so much better in rendezvous and docking these day than we were during Apollo flights, the reliability is so much higher - just take a look how many Falcon-9 flights in a row are successful - so I don't think operationally LEO refuelling will present a significant problem. And I'm sure we need maybe a couple of years to see first examples of that.
Space is hard, yes. But we're getting better, for sure.