Soyuz reliability versus quality control

For American dudes in their twenties who follow space, the Soyuz rocket is the cockroach of the space age: ugly, ancient, and virtually unkillable. It first flew in 1966, before the moon landing, and it’s still launching today. On paper, that’s a triumph of engineering. But look closer, and the Soyuz’s legendary reliability tells a different story about the agencies that run it—a story of stagnation masked by lucky survivorship bias. For Roscosmos, Russia’s space agency, the Soyuz isn’t proof of excellence. It’s a monument to a frozen industrial culture that can’t build anything new, while quality control crumbles under the weight of corruption and brain drain.

Let’s start with the numbers, because that’s what Americans respect. As of 2025, Soyuz has logged over 1,900 launches, with a success rate hovering around 97 percent. That’s solid. But that record is heavily weighted by early Soviet launches where “success” meant the rocket didn’t explode on the pad. Modern Soyuz launches—especially crewed missions to the International Space Station—have a perfect safety record since 1983. That sounds unbeatable. Until you remember that the rocket hasn’t fundamentally changed its design since the Brezhnev era. That’s not reliability through innovation. That’s reliability through sheer repetition, like a factory worker who never misses a day because he never gets a day off. The downside? Every Soyuz is a hand-built artifact, tweaked by craftsmen who learned their trade in the Soviet system. There’s no modern manufacturing digitization, no rigorous automated testing that would catch problems before they become disasters.

And disasters have happened. In 2018, a Soyuz capsule carrying NASA astronaut Nick Hague and cosmonaut Alexey Ovchinin suffered an abort two minutes after launch. The crew survived, but the investigation revealed a sensor that had been damaged during assembly—by a worker who bent the part during installation. That’s not a design flaw. That’s a quality control failure by a human being. The root cause traced back to a system where workers are underpaid, under-trained, and pressured to meet quotas. Russia’s state-run defense industry has been bleeding expertise for decades. Young engineers don’t want to work for a government paycheck when they can code for Gazprom or emigrate to the West. The guys left building Soyuz are aging out, and their replacements are cutting corners.

This is where the “agencies” part of the equation bites. Roscosmos is not NASA. It’s a state corporation that answers to the Kremlin, not to independent oversight or public accountability. When a Soyuz fails, the agency doesn’t issue a transparent, televised press conference with engineers pointing to CAD models. They bury the report, blame the component supplier, and move on. The lack of a watchdog culture means that quality control is driven by informal networks of “it worked last time” rather than rigorous statistical process control. In the United States, SpaceX runs thousands of tests on a single Raptor engine before it ever sees a launch mount. In Russia, they run the engine on a stand, and if it doesn’t explode in thirty seconds, it’s good enough. That “good enough” attitude has worked for fifty years. But it’s a house of cards.

The real problem for Roscosmos isn’t the Soyuz rocket. It’s that the rocket is the only thing they have left. The agency’s budget has been slashed since the Soviet collapse, and Putin’s priorities for space have shifted to military systems and prestige projects like the delayed Angara rocket. Meanwhile, NASA and its commercial partners are flying Crew Dragon and Starship, both of which are leaps ahead in terms of automated safety systems, reusability, and modern supply chains. Soyuz still flies because it’s paid for by international customers—mostly NASA astronauts who needed a taxi to the ISS. But those days are ending. With Dragon operational and Boeing’s Starliner limping along, the demand for Soyuz seats has cratered. Roscosmos is now flying cosmonauts to the station on Crew Dragon, which is a humiliating role reversal for a nation that once led the space race.

That leads to the bigger picture: the Soyuz’s reliability is a Russian cultural artifact, not a sign of agency competence. It’s a testament to a system that froze engineering in amber and then coasted on inertia. The minute a real problem emerges—like the 2018 abort or the 2020 coolant leak on a Soyuz capsule docked to the station—the agency struggles to respond. Their quality control methods are analog in a digital world. They rely on bureaucratic checklists and grizzled old hands, not AI-assisted diagnostics or real-time telemetry analysis. That worked when your only competitor was yourself. But when the competition is a private company that treats every launch as a data-generating experiment, the old guard looks obsolete.

For the casual space fan who just wants to know what’s next, here’s the bottom line: The Soyuz will keep flying for a while longer, because Roscosmos has no replacement ready. But its once-sterling reputation is already tarnished by near-misses and aging infrastructure. The agency behind it is a shrinking, risk-averse bureaucracy that has chosen to preserve a 1960s design rather than reinvent itself. Reliability isn’t the same as progress. And in the 2020s, progress is what matters. The Soyuz may be reliable, but Roscosmos is not.