Published on

SpaceX caught its own ship in mid-air, and that is what optimism looks like

Authors

Listen to this article · SpaceX caught its own ship in mid-air, and that is what optimism looks like

0:00 / 0:00

Watch this article · SpaceX caught its own ship in mid-air, and that is what optimism looks like

On 14 November, for the first time, SpaceX caught the upper stage of Starship — the ship, not the booster — as it came back from orbit. Ship 42 flew 14 times around the Earth, re-entered, and settled into the arms of the launch tower at Starbase, Texas, at 16:20 local time (04:20 on 15 November, Hanoi).

The booster did not make it. Super Heavy could not be caught on this flight, so it was diverted into the Gulf of Mexico and detonated before hitting the water. Fifteen flights in, the company has now recovered each of the two halves by tower — just never both on the same day.

15
test flights of Starship to date
14
orbits completed by Ship 42 before the catch
11,000+
Starlink satellites now active in low Earth orbit

What happened, in order

  1. Ascent

    Starship lifted from Starbase on 13 November local time; both stages separated on plan and headed for space.

  2. The booster is lost

    Super Heavy could not be caught, so it was steered into the Gulf of Mexico and destroyed before impact.

  3. The ship keeps going

    Ship 42 reached low Earth orbit and circled the planet 14 times, also deploying 26 next-generation Starlink V3 satellites.

  4. The catch

    It re-entered, came down through the cloud, and was taken by the chopsticks on the launch tower — a manoeuvre the booster has managed three times, and the ship never until now.

The hard half is the ship: it arrives with orbital velocity, not a suborbital arc.

Dan Huot of SpaceX called it "the missing link" during the broadcast, and described the manoeuvre as falling from altitude, through a wall of fire, into steel arms. The company played "Baby Come Back" as it came down.

Why the ship matters more than the booster

A booster flies a suborbital arc and comes back quickly. A ship has been to orbit, so it returns carrying far more energy — which is why catching it took fifteen flights and a booster catch took three.

Full reusability is the whole commercial argument. SpaceX needs Starship to keep adding capacity to Starlink, which is the business actually paying the bills, and NASA's plan to move crews and cargo between lunar orbit and the surface depends on the same vehicle. The Mars ambition sits behind both.

  1. Capacity for Starlink

    More satellites per launch, which is the business actually paying for the programme.

  2. Lunar cargo

    NASA’s plan to move crews and freight between lunar orbit and the surface assumes a vehicle that can land and return.

  3. The cost curve

    Reuse only pays at cadence: one catch is a demonstration, fifty a year is infrastructure.

What full reusability is for — in the order the money arrives.

The week's other story, and the one we would rather report

The same week brought a very different kind of technology headline: the uncertainty in the job market, with AI systems now doing work that people were told to train for, and companies announcing restructurings in the language of efficiency.

We do not think those two stories cancel each other, and we do not think the second one is the whole picture.

Amid the uncertainty in the job market due to AI, we are hopeful for the future with science and engineering advancement, with the help of AI. A ship caught out of the sky is what that hope looks like when it is made concrete: thousands of people, decades of iteration, failures that were reported rather than hidden, and a machine that does something no one had done before.

AI is part of how that happens now — in design, simulation, testing and analysis. It is also, on its own, not a reason for confidence. The confidence comes from what people build with it.

The honest caveat is that one catch is not a routine. The booster failed on this flight, the vehicle still has to fly often enough to be trusted, and the launch cadence that makes reusability pay has not yet happened.

What to watch

Whether the next flight catches both halves in one attempt, whether the cadence reaches the point where a catch stops being news, and whether the lunar variant stays on NASA's schedule. Those three would move Starship from a demonstration to infrastructure.

Source: VnExpress, 15 November 2026, reporting Space and CNBC. Flight details, the booster loss, the 14 orbits, the 26 Starlink V3 satellites, the 11,000-plus constellation and the vehicle specifications are as reported there. The SpaceX quotations are from the live broadcast as quoted in that report. The last section is our own editorial position, not a claim from the reporting.

Contact: [email protected]