At least 126 technicians and cosmonauts were killed by accidents during the Soviet space program. It’s rumored that additional cosmonauts died on missions that the Soviets covered up.

We also lost all three astronauts in a test of the Apollo capsule which doesn’t count as flight but I think we would agree still counts in the buckets of blood both programs have spent.

If you do more tests which kill more people before declaring the system good, you don’t get to not count those people like it’s a different color of money in some bureaucratic dystopia.

As far as technicians, I've always felt sad for the two Space Shuttle technicians who died in 1981 after entering a nitrogen-filled engine compartment (John Bjornstad and Forrest Cole).

It's very sad but it's not a bad way to go. I was a nuclear plant supervisor for about 20 years. One day the temp started to increase in the control room. The 3 AC units were behind the control room so I headed there to check on things. Another guy said he'd go with me. These things are as big as locomotives. Well while we were in the room a refrigerant relief valve lifted with a huge roar. Turns out, amazingly, the relief was designed to vent into the room. We made it to the card reader to get back into the control room but my card kept being rejected. I felt myself sliding down the wall and next I knew I was sitting in a chair in the control room feeling better and better. If my buddy hadn't decided to come with me I would have died in that room.

There was no panicky brain screaming "we need air" or any feeling of being O2 short. It was just normal breathing. If you're going to end it all that's the way to do it.

I wrote an incident report with the obvious suggestion to vent the reliefs outside. It was modified the next outage.

"But what if we vent the toxic gas into the place where you go to see if there's a problem with the toxic gas..."

Three died in that accident. The third lingered for a couple years before passing.

That's likely to include the Nedelin catastrophe ( https://en.wikipedia.org/wiki/Nedelin_catastrophe ) -

"The most recent estimated death toll, released by Roscosmos on the 50th anniversary of the accident and originating with agency engineer Boris Chertok, was that 126 people had died, but the agency qualified the number by saying that the actual number could be anywhere from 60 to 150 dead."

The Nedelin catastrophe was an ICBM test run of a rocket with hypergolic fuel that exploded on the launch pad; there was the 1980 launch pad explosion of a (military) sattelite launcher that killed 44 people and injured 43. The Russian version of the article says that hydrogen peroxide filters were produced with catalytic materials, this was later fixed after another launch almost failed in the same manner (it doesn't say how that blast was avoided), initially they suspected an action of the ground crew to have causes the blast in 1980. Both blasts were kept secret until Perestroika in 1989. both explosions were on rockets with hypergolic fuel.

The Russian wikipedia article also has a list of 'other failures' - these are fires in an ICBM silo in Russia in 1960 - eight dead; the US had a fire in a Titan II bunker in 1965 that killed 53 and another one in 1980 that killed one man; also Brazil had a fire of a liquid fueled rocket that killed 21, also on start preparations. (and another blast in Plesetsk in 1973 that killed seven http://www.plesetzk.ru/index.php?p=1973&d=doc/disaster - wikipedia doesn't list everything)

Rocket fuel can be very dangerous stuff! At least liquid methane/oxygen aren't hypergolic.

https://en.wikipedia.org/wiki/1980_Plesetsk_launch_pad_disas... https://ru.wikipedia.org/wiki/%D0%9A%D0%B0%D1%82%D0%B0%D1%81...

I've always loved Ignition![0] for the perspective it gives on working with the chemistry of this stuff.

[0] https://www.amazon.com/Ignition-Informal-Propellants-Univers...

But the U.S. has sent 3x as many citizens into space: https://en.wikipedia.org/wiki/List_of_space_travelers_by_nat...

So combining with your statistic the U.S. has a fatality rate of 4.4% versus Russia's 3.3%. Given the low N it's probably within margin of error.

At least 60 of those Americans were sent up on the Russian Soyuz: https://en.wikipedia.org/wiki/List_of_Russian_human_spacefli... Also, How are you doing the math to get a 4.4% fatality rate? The numbers are much lower than that.

Soyuz has launched 394 people and killed 4 of them. The most recent fatality was 50 years ago. The Shuttle launched 833 people and killed 14 of them, most recently in 2003 (8 years before the program ended). That gives Soyuz a fatality rate of 1.0% and the Shuttle a fatality rate of 1.7%. In other words: You're 70% more likely to die if you ride the Shuttle instead of Soyuz.

The more you look into the specifics, the safer Soyuz seems. The Soyuz deaths were early in the program while the Shuttle deaths happened when the program was mature. (Soyuz 11 was the last fatality and it was the 10th manned mission. STS-51-L was the 25th Shuttle mission and was considered to be safe enough for a civilian teacher to ride along. Oops.)

Soyuz's design is inherently safer. It has a launch escape system (which saved the crew of Soyuz 7K-ST No.16L).[1] The Shuttle did not. Soyuz's heat shield is much more robust than the Shuttle's. Soyuz's crew section is on top of the rocket, reducing the chance that any falling debris (or shrapnel caused by an explosion) will harm the crew module. Lastly, Soyuz uses liquid fueled rockets that can be shut off at any time. The Shuttle's solid boosters could not be throttled or shut off. All of these design decisions make for a simpler and safer vehicle. If I had to pick one, I'd ride Soyuz for sure.

1. https://en.wikipedia.org/wiki/Soyuz_7K-ST_No.16L

Of course proving a simple design and sticking with it is safer! Not going to space and being content with remaining on the ground is safer still! Water is wet! The sky is blue!

We can all see what happened: the soviets stopped spending money on innovation and tried to re-frame their stagnation as operational brilliance. Sure, Soyuz has a reliability niche, and I'd rather ride on it too, but I'd rather fund efforts to advance space exploration, even if they wind up being a clusterfuck like the shuttle.

Aren't solid boosters, like those used by the Shuttle, a simpler design than liquid engines?

On top of that, the Raptor engines used in the Starship, which seem to be considered state-of-the-art, are based on a soviet design from the 60s.

Raptors are state of the art. They are based on a soviet design from the 60s in the same way that they are based on US designs from the 60s.

It's weird that people seem so keen to talk up the soviet part and talk down the US part.

I think he is referring to the Soviet https://en.wikipedia.org/wiki/RD-270 full-flow staged combustion engine from the 60s. The Americans had not been able to preserve the high-pressure hot oxygen from eating the preburner.

I don’t think it makes an enormous difference, but not all of those US citizens flew in US rockets (nor did all Russian ones)

Also, quite a few of those did multiple flights. Looking at that list, I think that moves the needle in favor of US rocket safety.

Soviet spaceflights were quite long compared to American - Salyut-6 and Salyut-7 both had more expeditions than Skylab, and that's before Mir, while Shuttles were limited to about a month in space per flight. Of course launch and landing are more dangerous phases of flight - Apollo-13 notwithstanding - so it makes harder to pinpoint an objective measure for safety.

Both had two vehicles lost in space/reentry, the shuttles just held more astronauts. In terms of percentage of fatal missions the two were roughly equivalent, with Soviet failures front-loaded into less sophisticated and less safe early vehicles while the US failures were later on an overly-complicated system design.