SpaceX catches a Starship upper stage with its launch tower for the first time, a key step toward full reuse
What happened? On Saturday 14 November 2026, SpaceX caught a Starship upper stage with the arms of its launch tower for the first time. Ship 42 came back to Starbase, Texas, after about 22.5 hours and 14 orbits on Flight 15, lit three Raptor engines for a landing flip and settled between the "Mechazilla" chopstick arms at about 4:20pm ET (22:20 Amsterdam time), according to SpaceX and Space.com. Earlier in the flight it deployed 26 Starlink V3 satellites. The tower had caught Super Heavy boosters three times before, but never the ship, which returns from orbital speed and is the harder target.
The booster: The plan was to catch both stages. Booster 22 cut its boostback burn short on a low fuel mass alarm, and its guidance, navigation and control system aborted the catch and diverted it to the Gulf of Mexico, where the flight termination system was triggered. SpaceX says the root cause is under investigation, per Aerospace America.
Why the catch matters: SpaceX calls ship recovery a critical step "in making Starship a reliably, rapidly, reusable rocket". Catching the ship means it needs no landing legs, wings or wheels, so that mass can go to payload, and it comes down on the same launch mount it will fly from again. The Space Shuttle, by comparison, threw away its external tank on every flight, and its boosters and orbiter needed months of rework between missions. SpaceX's stated goal, in the words of its Kate Tice, is "to immediately land the rocket and launch it again, just like an airplane." It is also the first time SpaceX has a whole Starship heat shield, about 18,000 hexagonal ceramic tiles, back on land intact to inspect.
What the mass to orbit is for: Starship V3 is designed to lift more than 100 tons to low Earth orbit. A rocket that can fly that much mass again and again spreads the cost of building it over many flights, which is what would push the cost per kilogram down. SpaceX needs that cadence for three things: deploying Starlink V3 satellites at scale as it plans to retire Falcon 9, refuelling Starship in orbit so NASA's Artemis lunar lander can reach the Moon, and launching the Starmind orbital data centre satellites it is developing for AI compute, with prototypes planned for 2027, per Ars Technica.
How often Musk wants to fly: Elon Musk has given several cadence targets this year. On 19 February he posted on X: "It will get really nutty when Starship is launching every hour in 3 years." On SpaceX's first earnings call on 4 August he said: "We expect the cadence of flights to be increasing rapidly, and, probably a year from now, we will be doing at least one flight a day, possibly more." On 21 August he posted: "We're aiming to reach 30+ Starship launches/day in 2030, which is ~10k annualized." On the third-quarter call on 20 October the near-term target was lower: "We expect the Starship to fly weekly in the first half of next year," with twice-weekly flights in the second half of 2027, per The Wall Street Journal and Ars Technica.
What's still unproven: No caught ship has flown again yet. SpaceX president Gwynne Shotwell said the company would try to refly a Starship upper stage early in 2027, alongside a ship-to-ship refuelling test. Flight 15 was Starship's fourth launch of 2026, and Starbase is currently authorised for 25 launches a year, per Space.com.
Sources: SpaceX: Starship Flight 15
Aerospace America
Space.com: Flight 15 live
Space.com: Why Flight 15 matters
Space.com: August earnings call
BBC News
The Verge
Ars Technica
The Wall Street Journal
Benzinga: Musk on hourly launches
Benzinga: Musk on 30+ launches a day
