State of the art right now is capsules that rely on parachutes that have no go-arounds, no ejection seats and no aerodynamic ability to correct.

edit: And the Space Shuttle which had go-around and aerodynamic ability to correct was even LESS safe than capsules.

Unlike the Soviet Buran Shuttle, there was no option to install an auxiliary jet engine and you had to be a very good pilot to compensate for the Shuttles brick like gliding qualities.

Starship actually _could_ have go-arounds, depending on how much fuel margin they have available.

The space shuttle?

The space shuttle landed aerodynamically. Starship would be taking on the same risk profile of a launch system with no escape mechanism, and reentry based on ablative tiles, without the one redeeming feature the shuttle did have which is the ability to control its' own descent unpowered and select a suitable landing site.

Starship could conceivably land on anything reasonably flat if it needs to land in an emergency compared to Super long special runway the Shuttle needed.

I believe they can just burn extra fuel margin to do a traditional vertical landing though. It costs payload, but it should be doable.

When maximizing cargo delivery to the surface of Mars, I'm sure they'll always do the flip.

I agree it'll probably be quite a while before they reach that level of safety though. This is brand new technology, and it's not going to reach the level of reliability of traditional aircraft without decades of experience with different failure modes and techniques for mitigating them.

By contrast you have a fatal airplane crash in the order of 10,000,000s flights.

Space fight is a lot more dangerous than normal in-planet transportation, and the Space Shuttle is on the more risk side of it (but any project has too small numbers to generalize).

Planes glide, and are generally only unsafe during a few critical moments after tire-lift.

Planes that don't glide are generally considered unsafe and highly specialized.

I fail to wrap my head around how a mechanism that uses active propulsion for all phases of flight could ever match the safety of a mechanism that exploits passive physics in the face of active propulsion system-failure.

That said, I couldn't tell you how to achieve space-flight otherwise except for maybe a shuttle-lander type system -- I am simply stating that the comparison of risk between air travel and space travel seems absurd to me.

Of course space travel would be more dangerous.

I get your point but things become normalised over time and mechanisms/technology invented to deal with the risk and conditions. If you'd said to people 150 years ago that there would be (well pre-covid days) at any one time about a million people at 30k feet flying around in basically a tin can, journeying across the world in a day, they'd say you're mad and it's too dangerous to achieve. They didn't have the technology capable, like a jet engine or fabrication techniques. Maybe in 50 years we do have a new technology that meets the crtieria needed, maybe not, but the same happened before and I'm sure it'll happen again.

So, rockets have had almost a century of development too. You could even stretch that and claim more like 6-700 years, as the Chinese had "one of the world's earliest multistage rockets and ballistic cruise missiles" around the 14th century: https://en.wikipedia.org/wiki/Huolongchushui

The flip-and-land maneuver doesn't permit much "uh oh, it's gone wrong..." time to correct, either. I think they're going to have to go more in the direction of airliner-style "look, things kinda have to work" - there's no ejection seat, no parachute.

If humans were involved in the landing I think the design would be significantly different.

With production legs instead of the tiny stubby test ones, I suspect it'd have worked.