But three of the missile’s 33 booster engines did not start on launch, and several stopped working as the missile climbed higher into the air. About four minutes after takeoff, the missile began to fall violently and exploded over the bay.

As part of the new space race of the 21st century, the failure of Starship’s first flight drew worldwide attention.

SpaceX said its failure provided several lessons to be learned from the next launch, which won’t take place until weeks after the damage to the Boca Chica base is cleared. However, SpaceX does not see this process as a failure, but as a starting point to solve the problem before the next launch.

But failures with bangs and bangs often result in better pictures and news than conventional successes, and they often receive more media coverage. But more than 75 years since mankind began sending rockets into space, how likely is a successful rocket launch?

David Wade, a risk assessment specialist at Atrium Space Insurance and an underwriter supporting commercial space launches, says: “2022 saw 186 launches. Those satellites carried 2,509 satellites into orbit, mostly from Starling, the satellite platform for SpaceX.

Wade adds: “A Falcon 9 rocket launches 60 satellites simultaneously, and only eight of 186 launches fail.”

The failure rate is about 4 percent, once in every 25 launches. But Wade says 2022 is an exception.

Last year saw a record number of missile launches. This reflects how rapidly the space movement is growing, with more than 40 operations expected by 2021, more than double the number seen five years ago in 2017. But this increase in mobility also means an increase in risks.

“Last year we had more failures than some years because we’re starting to see a lot of new launch vehicles, and the first launches of launch vehicles are always more complicated,” explains Wade.

Starship and NASA’s Artemis program, which successfully launched for the first time in November 2022, are the most popular of the new launch vehicles, but they’re just the beginning. New launch vehicles come with a lot of uncertainty, especially on their first flights.

SpaceX’s Starship lifts off from the launch pad, but explodes minutes into its first flight.

“Typically, on the first or second launch, we expect a failure rate of up to 30 percent, then things start to pick up, and by the tenth flight, we expect the failure rate to drop below 5 percent,” Wade says.

“It’s very difficult to determine why some people fail and others fail,” he says. “In the first two or three launches, a lot of effort goes into doing massive research, improving quality control, making sure everything is right. , we’re still there, and we usually see failures during the first two flights when unexpected things happen.”

“This low failure rate is because the Russian Soyuz rocket has been in use for decades, since Sputnik in 1957,” said David Todd, head of space content at Ceradata, which has researched satellites.

Dodd says that manned spaceflight is on high alert, explaining that the major loss of repatriating Space Shuttle Columbia in 2003 was due to a “launch failure”; Damage to the thermal protection system occurred during launch, and manned spacecraft are equipped with safety systems that can save the crew if the launch pad fails, as happened with Soyuz 18A in 1975.

By 2022, SpaceX will launch a successful Falcon 9 rocket every six days. China also broke its record, launching 64 B missiles, only two of which failed. Even New Zealand, not exactly famous for space, managed to launch nine rockets from a burgeoning space launch center on the east coast of the North Island.

This new boom presents interesting challenges in the relatively new field of aerospace insurance. Wade used the upcoming European Space Agency’s Ariane 6 missile as an example: “The insurance process starts with the third flight of the missile, and it’s hostage to the success of the first two launches of this missile, otherwise it’s out there. The purpose of the insurance.”

Advanced computer modeling and years of experience launching missiles into orbit mean more accurate predictions, but each new launch system is subject to unknown decisions that remain speculative until launch, Wade says.

“There are hot, high-pressure gases, environments that change very quickly. And if something goes wrong, there’s not much you can do to prevent it.”

There are companies that have tracked space launches for the past eight decades and are capable of depicting the likelihood of failure. Each week, new releases are added to Seradata’s database, which goes out to clients of the insurance companies Wade works with.

Ceradata’s failure data relates to not only the rocket failing to reach orbit, but also the payload not reaching the intended orbit or sustaining significant damage during launch.

For example, launches in the 1950s, which saw the start of the space race, had a staggeringly variable failure rate of more than 70 percent. By the early 1960s, that number had begun to decline and hovered around the same point at 7 percent (9 percent if satellite launch failures were included).

“Early in your career you always have problems, big failures, and over time you get comfortable with Western launch vehicles,” says Todd.

However, the same cannot always be said about missiles manufactured in countries like Russia; The Proton Launch System, which sent the Soyuz spacecraft to the International Space Station, partially bucked this trend.

The first launch of the Proton rocket in 1965 was a stunning achievement.

Todd compared the Proton’s success rate to that of the European Ariane 4 and the US Delta 2, both of which achieved more than 100 successful launches respectively.

“The Proton failed every 20 or 25 flights. I think a lot of people got into the industry because of the space race in Russia in the 1960s, and they were elite university graduates.”

Wade added that the collapse of the Soviet Union greatly reduced the Russian space industry, leaving it with fewer material resources.

“A lot of this knowledge is lost. In the case of the Proton, anyone involved in building its components knows more than just the instructions provided, which can make the system work better with simple touches.”

When it comes to the first launch of new missiles, even after eight decades have passed since the beginning of the rocket age, the failure of the Starship seems very reasonable.

“Evolution is one of the problems facing the missile industry, it starts out going through bad phases, until it reaches a good development, the missile designers are asked to build a better missile, and they start again.”

However, Wade discovered something else from the data, namely that the probability of failure increases dramatically with the sixth launch.

“We found two failures in the first 10 flights, and the first and second failure rate is typically 30 percent. Then the rate slopes downward.”

“Sixth stage often has a high failure rate. At that stage it is difficult to know the exact cause, getting used to the regular production cycle, or quality control is not yet up to the required level, or there are finishing touches and other things that lead to the failure of the launch process.