Tools chosen: kalshi_direct kalshi_related polymarket_related claude_news gdelt_news wikipedia code_execution

Sub-questions (Fermi decomposition)

Planner reasoning

This is a long-horizon technology-milestone question where the base rate is extremely low: SpaceX has not yet flown Starship to orbit with reuse, done orbital refueling, or landed anything on Mars. The key research is the current Kalshi price plus concrete program milestones (uncrewed Mars window attempts in 2026/2028/2029, HLS lunar crewed timing, Starship V3 flight status) to confirm the near-zero estimate.

# Current state Kalshi's own market (STARSHIPMARS-29DEC31) prices a crewed Starship Mars mission before 2030 at just 10% YES, down from ~16-23% thirty days ago. As of mid-to-late 2026, Starship has flown 13 times (8 successes, 5 failures), V3 has debuted but orbital propellant transfer is still undemonstrated, no uncrewed Mars mission has launched, and even the technically simpler Artemis III crewed lunar mission has been descoped to an Earth-orbit-only docking demo, slipping to late 2027. # Timeline of key events - 2016-2024: SpaceX repeatedly announces Mars cargo-landing targets (2018→2022→2026→2028), none met (confirmed pattern per scienceblog.com/multiple outlets). - 2024: Musk states 2026 uncrewed Starship Mars mission target, crewed flights "as early as 2029" (reported). - 2025: Starship V2 flies 5 times with mixed results (3 failures, 2 successes); Flight 11 (Oct 2025) closes out V2 (confirmed, Wikipedia/claude_news). - 2026-02: Artemis III role restructured; mission redefined as Earth-orbit rendezvous/docking demo rather than lunar landing (confirmed, Wikipedia). - 2026-05-22: Starship V3 debuts (Flight 12), partial success — one Raptor engine failure, booster missed boostback (confirmed, multiple sources). - 2026-05: NASA/media report Artemis III slipping to late 2027 (confirmed, space.com, twistedsifter, ecoportal). - 2026-07: Flight 13 tests V3 Starlink deployment, not propellant transfer (confirmed, space.com). - 2026-08: SpaceX website now lists Mars cargo flights "no earlier than 2028"; commentary describes 2026 uncrewed Mars attempt as effectively shelved (reported, wlockett/Medium, scienceblog.com). - 2026-09 (planned): Flight 14 (3rd V3 flight), destination/purpose unspecified (reported/rumored). # Event Will SpaceX launch a manned Starship mission to Mars before Dec 31, 2029? # Outcomes to forecast Yes / No # Kalshi market anchor STARSHIPMARS-29DEC31 current YES price: **10%**. 7-day change: 0%. 30-day change: **-6%** (declining trend). Price range over 61 days of data: 8%-23% — market has drifted down from earlier highs. Average daily volume: 225 contracts (moderate liquidity). # Sub-question answers 1. **Current price/90-day move** — 10% YES currently; down from a high of 23% in the observed window, 30-day change -6%, 7-day flat (kalshi_direct). 2. **Mars transfer windows before 2030** — Two remain: late 2026 and late 2028/early 2029. SpaceX has NOT committed to a crewed 2026 attempt (uncrewed only, and even that has reportedly been shelved); website now targets uncrewed cargo landings "no earlier than 2028" (claude_news, code_execution). 3. **Prerequisite capabilities** — Orbital propellant transfer NOT yet demonstrated as of mid-2026; V3 just debuted (May 2026) with partial success (engine-out, missed boostback); full reusability partially shown (booster catch demonstrated on Flight 5) but reliability still maturing (Wikipedia, claude_news). 4. **Artemis III status** — Slipped to late 2027, restructured from a lunar landing to an Earth-orbit-only Orion/HLS docking demo; actual crewed lunar landing now targeted for Artemis IV (~2028) (Wikipedia, multiple news sources). 5. **Musk's recent statements** — Musk cited "50/50" odds on 2026 uncrewed mission; crewed flights suggested "as early as 2029" if uncrewed 2026 succeeds; historically, every SpaceX Mars deadline since 2016 has slipped 2+ years (scienceblog.com pattern; claude_news). 6. **Blockers** — Orbital refueling undemonstrated is the primary technical bottleneck; ECLSS/life-support for 6-9mo transit not yet certified; no formal NASA/SpaceX program commitment to crewed Mars flight in this decade (code_execution estimates). 7. **Cross-market agreement** — Related Kalshi markets price uncrewed Mars landing before 2030 at 21% (higher than crewed at 10%, as expected), Musk visiting Mars in his lifetime at 11%, human Mars landing before 2050 at 26-37%. No Polymarket markets found on this topic (0 matches). # Key facts (high-confidence, factual) 1. [Wikipedia] Starship has flown 13 times as of July 2026 (8 success, 5 failure); no orbital propellant transfer yet performed. 2. [Wikipedia/NASA] Artemis III redefined as Earth-orbit docking demo, not a lunar landing; slipped to late 2027. 3. [claude_news] SpaceX's website Mars cargo target is now "no earlier than 2028," down from earlier 2026 claims. 4. [kalshi_direct] Market YES price 10%, trending down over 30 days. # Cross-market signals - Kalshi related: uncrewed Mars landing before 2030 (KXSPACEXMARS-30) = 21%, down 7% in 30 days; Elon-visits-Mars-lifetime = 11%; humans-colonize-Mars-before-2050 = 22%. - Polymarket: no active matching markets found. - Sportsbook implied: N/A (not applicable to this event type). # Analyst opinions and speculation - code_execution Fermi model: multiplicative chain of required milestones (propellant transfer, uncrewed landing attempt+success, crewed-lunar precursor, life-support, cadence, formal commitment) yields ~0.3%-5% central estimate, best guess ~1-2%. - Commentators (scienceblog.com, Medium) frame SpaceX's Mars timeline as a decade-long pattern of missed deadlines, suggesting continued slippage is the base case. - Musk's own "50/50" framing refers only to the uncrewed 2026 attempt, not crewed missions — an even lower-confidence crewed timeline follows structurally. # Directional lean per outcome - **Yes**: Supported only by Musk's aspirational statements and SpaceX's rapid iterative test cadence (100+ launches/year across programs); no uncrewed Mars mission has flown yet and crewed Mars requires it to succeed first, then a second landing window in 2028-29. - **No**: Strongly supported by undemonstrated orbital refueling, Artemis III's slip/descoping (removing the crewed-landing precursor), 2026 uncrewed Mars attempt reportedly shelved, and SpaceX's decade-long historical pattern of 2+ year slips on every Mars deadline. Kalshi and related markets (10%, 21% for easier uncrewed version) both price this as unlikely. # Gaps / unknowns - No confirmed formal decision on whether SpaceX/NASA will attempt any Mars mission (crewed or uncrewed) in the 2028-29 window. - No data on ECLSS certification timeline or FAA/planetary-protection regulatory status specifically for crewed Mars. - Polymarket has no comparable market to cross-check. # Calibration anchors - Kalshi current YES price (anchor): **10%**, down from 30-day-ago ~16%. - Related Kalshi uncrewed Mars landing (easier bar): 21%. - Historical precedent: SpaceX Mars deadlines missed by 2+ years in 2018, 2022, 2026 cases — supports continued downward pressure on this market's price toward single digits.

1. Both forecasts land below the Kalshi anchor (10%) and even below the Fermi model's stated range (0.3%-5%, best guess 1-2%), yet neither justifies why market-implied probability should be discounted rather than treated as a floor—longshot-bias arguments are asserted but not quantitatively supported, and both may be over-anchoring to the Fermi estimate rather than the live market price. 2. Neither forecast addresses the cross-market inconsistency flagged in the brief: uncrewed Mars landing before 2030 prices at 21% while "human Mars landing before 2050" prices at 22-37%, which is barely higher than the 2030 uncrewed figure—this suggests the market itself may be inconsistently priced, and neither forecaster flags or exploits this as a reason for caution about anchoring too tightly to Kalshi. 3. Both forecasts treat the 2028-29 transfer window as essentially foreclosed by the shelving of the 2026 uncrewed attempt, but neither explicitly models the tail scenario where SpaceX abandons the "uncrewed-first" sequencing under Musk's known risk tolerance (Starship's own test philosophy is "fly fast, fail fast, iterate crewed-adjacent") — a low-probability but non-negligible scenario where a crewed mission launches without a prior fully successful uncrewed landing. 4. Neither forecast discusses regulatory/political risk in the other direction, i.e., a scenario where NASA or an external commercial customer forces a delay-inducing safety review after any high-profile Starship failure — both incorporate historical slippage into the "No" case but do not consider whether the current 5-failure track record (13 flights, 5 failures) could trigger additional FAA-mandated groundings that further compress the probability, meaning even the ~3-6% estimates might still be too high. 5. Forecast 1's confidence (0.87) and Forecast 2's confidence (0.72) differ substantially despite near-identical point estimates (3% vs 6%) and nearly identical rationales; this inconsistency in stated confidence isn't explained by any difference in reasoning, suggesting one or both confidence values are not well-calibrated to the actual uncertainty in their models. 6. Neither forecaster explicitly weighs the "Musk historically slips 2+ years" pattern against the fact that the remaining window (late 2028/2029) is itself already a slipped target from earlier 2026 promises — meaning even this window could itself slip past 2030 with high probability, which arguably should push both estimates toward the very low end of the Fermi range (0.3-1%) rather than settling near 3-6%.

3% (80% confidence)

Yes: 3% No: 97%

A crewed Starship Mars launch before 2030 requires a chain of unmet prerequisites: orbital propellant transfer is still undemonstrated, V3 only debuted in May 2026 with a partial success, no uncrewed Mars mission has flown, SpaceX's own site lists cargo landings 'no earlier than 2028,' and Artemis III has been descoped to an Earth-orbit docking demo slipping to late 2027. With only the late-2028/early-2029 window realistically remaining and a decade-long record of 2+ year slips on every Mars date, the physical/programmatic base rate is very low — the Fermi chain gives 1-2%. The Kalshi anchor of 10% (down from 23%, still falling) almost certainly embeds longshot bias and the residual option value of a 3.5-year horizon; the fact that the strictly easier uncrewed-landing market sits at only 21% implies crewed should be a small fraction of that, not half. The critique's points about ignoring an unconventional crewed-first gamble and possible market mispricing argue for not going to the extreme low end, while its points about further FAA/failure-driven slippage argue the other way — these roughly offset. I settle at 3%, below the market but above the pure Fermi central estimate.

Total pipeline time: 180.9s