SLS can theoretically launch 100T to LEO. Starship can theoretically launch 150T reusable, 250T expendable. Construction rate is already on the order of around one per month, although obviously they are working out issues with the launch infrastructure to actually be able to make that many launch attempts.
That depends on your design and if you need experimentally derived inputs
When half your components have decades of reliability under their belt of course it’s easier to have a successful first launch.
If your development methodology results in a product that is unsustainable to launch due to costs and scalability, that absolutely seems relevant to whether that methodology was successful.
SLS is going to be tossing four RS-25 engines per launch (with *each* engine costing the same as ~2-3 entire Falcon 9 launches) into the ocean as the planned course of action. That is by design - and not some accident during a test flight.
Sure, RS-25s were designed to be reusable. But it's not like they were being used for anything; they were just hanging out in a warehouse since the Shuttle days. If NASA was going to put them in a museum or something it would have already happened.
At least this way, they get used for what they were designed for.