§1Getting there
Starship’s flight record, the refueling gate nobody has passed, and what NASA and China are actually doing.
Two dates matter more than any other in the Mars conversation as of mid-2026, and neither has happened. The first is the day a Starship reaches Earth orbit. The second is the day one Starship pumps propellant into another while there. Everything downstream waits on the second: the Artemis lunar landing, any uncrewed Mars shot, every crewed date anyone has proposed. This chapter is about the distance between the flight record and the calendar.
The 2025 flight record
§1.1 2025 was a punishing year for the redesigned “Block 2” Starship. The Super Heavy booster proved reliable; the upper stage kept failing, until it didn’t. The table lists what actually flew.
| Flight | Date | Ship | Outcome |
|---|---|---|---|
| 7 | 2025-01-16 | Ship 33 | Lost before engine cutoff: harmonic vibrations caused propellant leaks and fire in the aft section; debris over the Turks and Caicos. |
| 8 | 2025-03-06 | Ship 34 | Lost to a similar engine-section fire. |
| 9 | 2025-05-27 | Ship 35 | First reuse of a Super Heavy booster. Ship got past cutoff, then lost during reentry to a methane leak; the booster was also lost. |
| — | 2025-06-18 | Ship 36 | Exploded on the Massey test stand during propellant loading. |
| 10 | 2025-08-26 | Ship 37 | Full suborbital mission and Starlink-simulator deploy. established |
| 11 | 2025-10-13 | Ship 38 | Last Block 2 flight; succeeded and closed out Pad 1. established |
Read the table both ways. Three ships lost in flight and one on the stand is a bad year for an upper stage. Two clean missions to close it out, on boosters some of which reflew, is the kind of recovery an iterative program is supposed to produce. The record supports neither the “Starship is failing” headline nor the “Starship is ready” one.
The gap that matters: orbit, then refueling
§1.2 Per Spaceflight Now (July 25, 2026), “no Starship of any kind has yet been put into Earth orbit,” and the lunar-lander variant “has not yet flown.” Independent trackers place the first orbital attempt (Flight 14, with the Version 3 vehicle and Raptor 3 engines) around Q3 2026.
The orbital propellant-transfer demonstration is the enabling technology for both Artemis and Mars. It is targeted stated: NET (no earlier than) late 2026 and had not occurred as of August 2026. There is no way around it: a Mars-bound Starship must be refilled in orbit, and the stated 2026 plan requires on the order of 40–60 tanker launches to fuel up to five ships, a cadence never before achieved by anyone.
The timeline against the track record
§1.3 Aerospace America’s documented history runs: a 2016 commitment to a 2018 Mars mission (a Dragon), then 2022, then a 2024-era target of “uncrewed 2026, crewed 2028–2030.” At the May 29, 2025 Starbase presentation, Musk said:
“I think we’ll probably have a 50-50 chance right now because we’ve got to figure out orbital refilling in order to have enough capability to go to Mars. But if we achieve orbital refilling in time, then we will launch the first uncrewed Starship to Mars at the end of next year.”
The stated plan: stated goal up to five uncrewed Starships in the Nov–Dec 2026 window, carrying Tesla Optimus robots. If refueling is not mature, he has said he would slip to the next window (~2028–2029). That is a corporate aspiration, and this site tags it as one. The consistent independent read is that the 2026 uncrewed window is highly unlikely to be met on schedule while orbital flight and refueling remain undemonstrated, and that the crewed timeline is optimistic by years, mirroring roughly a decade of prior slips.
NASA and the United States
§1.4 The FY2026 budget process was turbulent. The White House proposal (May 2025) cut about 24–25% (~$6 billion) from NASA overall, gutting science, while adding ~$1 billion for “Mars-focused programs,” and called for canceling the SLS rocket, the Orion capsule, and the Lunar Gateway after Artemis III. In February 2026, NASA announced it was adding a mission and standardizing on SLS, targeting the lunar South Pole landing in 2028, with the Artemis II crew (Wiseman, Glover, Koch, Hansen) preparing for a crewed lunar flyby as early as April 2026. A Blue Origin rocket explosion in May 2026 raised concerns for Artemis III and the fall-2026 CLPS (commercial lunar lander) missions.
One decision is firmer than the rest. In December 2024, NASA made its first firm Mars-architecture choice: nuclear fission surface power for the first human Mars missions. established as a decision; the hardware is covered in §5 Life support.
China
§1.5 Tianwen-3, China’s Mars sample-return mission, is set for a two-launch campaign in 2028 with the return of at least 500 g of samples by about 2031 — potentially the first return of material from another planet. That would be ahead of the troubled NASA–ESA MSR (independent reviews flagged a cost above $11 billion and a return no earlier than 2033). China has stated an intent to send a first crewed Mars mission around 2033 and eventually build a base. stated goal
Expedition or settlement?
§1.6 Everything in this chapter is about getting there, an expedition question. Reaching orbit, refueling, and landing are engineering unknowns; the report’s own caveat is that these “will likely be resolved within a few years one way or another.” Nothing here bears on whether people can stay. That question begins in §2 and turns decisive in §4.