They're gathering a ton of data to make it robust! Many of these engineers built Falcon 9, and I have a pretty high degree of confidence they'll shake out the issues. SpaceX operates very differently from traditional aerospace, so we'll likely see many more issues come up before Starship is human rated.
Why? For NASA, human-rating - for either Earth launch or Earth re-entry - requires a very detailed engineering analysis of probability of loss-of-crew (LOC), and then NASA has some maximum LOC probability they allow (I believe 1-in-500, which is 0.2%). That analysis is based on engineering data, which can include simulations, data collected from actual flights, and data from ground-based testing. If SpaceX can demonstrate N successful uncrewed landings (with maximum G forces within acceptable limits for crew, etc), for sufficiently high N, logically the LOC probability will fall beneath NASA’s threshold, and then NASA will human-rate it. However, they don’t actually have to land it 500 times - all they need is an engineering analysis which calculates the LOC probability as being below threshold, and then NASA’s own engineers review it, and once NASA’s engineers are confident it is correct, the human-rating will be approved
Elon had that idea of using rockets for passenger transport so I’m sure it’s in the plan, but for astronauts I don’t see why you’d rush.
SpaceX wants NASA to human-rate Starship because their long-term plan is to retire Falcon 9 / Dragon.
How I expect it will happen: SpaceX will run their own internal analysis of loss-of-crew probability for launch and re-entry. No point going to NASA until they've convinced themselves they are going to meet the standards for human-rating. Once they believe they are, then they'll decide when is the right time to try to convince NASA of it too.
Obviously they are already doing this for lunar landing/launch. Some of that is going to be transferrable to Earth landing/launch. Other parts are unique – e.g. probability of surviving re-entry – but they need to estimate that anyway for non-crewed use cases.
That very analysis will include the fact that they have no launch abort system like normal capsule based systems and that the belly flop landing is much more dangerous than landing a capsule. Space Shuttle memories may come to mind. So the only realistic way to get the 1 in 500 confidence would probably be to really land hundreds of times for unmanned missions and not mess up even once. Granted, they may eventually get there, but I estimate this would take about two decades, if Falcon 9 is any indication.
My concern is that the flip maneuver is just too risky for landing with people on board even if Starship manages to do 200 perfect landings in a row (that would be a better record than the Shuttle).
You are talking about this as if it is based on feelings as opposed to probability calculations. SpaceX will present a probability calculation to NASA, who then either accepts it or disagrees with it on engineering grounds, not "memories". There is no formal requirement for a launch abort system in NASA's safety standards – just a requirement that the probability of death be below a certain threshold. A launch abort system is one way to get that probability below the threshold, but if you don't have one, that's okay so long as you have some other way of getting there.
> So the only realistic way to get the 1 in 500 confidence would probably be to really land hundreds of times for unmanned missions and not mess up even once.
What they do is list every possible failure mode, give it a probability of happening and probability of a lethal outcome if it happens, multiply those two probabilities together, and then add them all up. If the result is above the threshold, they will fail certification. If the result is below it, they then have to convince NASA engineers that (1) their probability estimates are accurate and have sufficient evidence to justify them (2) there are no failure modes they've omitted in the analysis. For (1), actual flight experience is a valid source of evidence, but there are others as well – such as simulations and on-the-ground testing of components. There is no minimum number of flights required to gather sufficient evidence, it all depends on how much non-flight evidence is available and how NASA engineering evaluates that non-flight evidence (which is going to depend on the component or failure mode we are talking about).
You are looking at this from a "big picture" perspective, whereas NASA will be looking at it scenario by scenario, component by component
Each lunar flight implies four launches and three landings on Earth, so the numbers should build up rather quickly.
In a lot of ways they will learn far more from the heat shield burn through around the flap(s) than they would have if they had been "lucky" and it had all gone perfectly.
You during testing you want things to fail, that is the point of testing. If it's all successful you only learn that under those conditions your design works, but if it fails, you learn another way to not do things.
When parts last longer than expected, it is considered something that needs "fixing". It is a signal that the part can be made cheaper, lighter, etc... If SpaceX had gone with heavy, thick tiles, and they did their job because they were overspecced, it is that much less payload capacity.
Even value engineering, which is often criticized when it comes to consumer products is a good thing. Yes, your new dishwasher is not as robust as the one made in the 50s, but it is also 10x cheaper (inflation-adjusted), and it can still wash dishes for maybe 10-15 years without repairs, at which point you may want a new one as technology has improved. Note that I am talking proper engineering, having a single point of failure that prompts a replacement is planned obsolescence and terrible engineering, there should be no single point of failure with good engineering.