ULA is pretty remarkable for it's run of new rockets not blowing up. Looking at ESA, JAXA, RosCosmos, ISRO, etc too is how I'm setting the par. A history like the Ariane 5 is pretty typical where flights 1 and 14 failed.
Yeah, 2 failures is par for OldSpace. NewSpace usually does much worse, though SpaceX's Falcon 9 and Rocket Lab's Electron managed to get the traditional par.
TBF: that wasn't an unsuccessful launch attempt, but a failure to not launch. Which affirms parent in that they seem to have work out all the kinks out during development.
Exactly as OP said, launcher failures happen and then you drive down their frequency.
Landing failures are still quite expected, especially on the first few tries. It's weird that they even tried on the first launch, but I don't even think of it as a try, I think of it as a "let's gather some data, and in the freakishly unlikely occurrence that everything goes perfect on the way down, we might as well load the landing software too".
I read about spaceship on one of their launches is that they attempted everything that it could possibly do on one of their boosters because you basically have the next iteration built so why not attempt anything for the telemetry.
Shuttle got very lucky. On the first flight, STS-1, an overpressure caused by the ignition of the SRBs forced the orbiter's body flap into an extreme angle which could have destroyed the hydraulic system controlling it. Had John Young know this had happened, he and Robert Crippen would have ejected, which would have destroyed the orbiter on its first flight.
There were only 2 ejection seats, enough for the crew of test flights but not the larger crew of operational flights.
The seats were only installed in Enterprise (the prototype, used only for suborbital tests) and Columbia (only enabled for STS-1 through STS-4 test flights, disabled for STS-5 the first operational flight)
The seats would only work at low altitude and speed (I've seen differing numbers cited). For the Challenger disaster they would've theoretically been useful (ignoring all the other factors), but they would've been useless for Columbia due to speed.
And it's not clear ejection would have actually been successful with the SRBs still active and right there.
As o11c mentioned, they only existed for the first few flights, the ones that only had two crew. It wasn't possible to have the election seats with the full shuttle crew so they were removed.
The ejection seats were essentially the same as those used in the SR-71, so they were survivable at shockingly high speeds and altitudes.
Norm is something like 3 rescheduling within a week from launch, 3 auto-aborts or equipment NoGo, 2 wayward boats, and 0.15-0.3 kaboom per launch. The fact that SpaceX haven't been letting wayward boats/planes for a while is remarkable by itself.
Isn’t the norm not crashing and succeeding? it’s only space x who normalized so many failures to “move fast”?