Skip to main content
MANIFOLD
Which Starship Variants will fly before 2028
19
แน€1kแน€4.7k
2028
87%
Tanker
59%
Propellant Depot
57%
Human Landing System (HLS)
37%
Other
34%
Chomper
17%
Mars Cargo
4%
Mars Crew
Resolved
YES
Pez Dispenser

Background

SpaceX is expected to develop several Starship variants for different mission profiles. Which will fly in the next three years (by 2028)?

Resolution Criteria

A variant will resolve as YES if at least one flight with the variant occurs before January 1, 2028. A flight is considered complete when the rocket leaves the launch pad, regardless of mission success.

Variant details:

  • Tanker: payload area designed to hold a liquid.

  • Human Landing System (HLS): Capable of carrying and landing crew. A test article with major systems that are missing or deliberately non-functional (including life support, airlock, etc) won't count. Does not need to have crew on board to resolve.

  • Propellant Depot: Configured to transfer and store propellant. Expected design features that separate it from a tanker include extra insulation and cryomanagement systems. Also expected to lack reentry hardware, but this is not a requirement. A non-starship-derived depot doesn't count (e.g. launched as payload instead of as a second stage).

  • Pez Dispenser: payload area designed to carry Starlinks

  • Chomper: a cargo bay door allowing a large payload to be released

  • Mars Cargo: capable of Mars entry, decent, and landing. Design must allow it to deploy payload to Mars surface.

  • Mars Crew: like Mars Cargo, but carries crew. Similar resolution criteria as HLS.

  • Other: something that doesn't neatly fall under existing categories. For example, a point-to-point variant that doesn't go orbital and lands on legs. Must have a real mission planned, rather than being a test-only stepping-stone towards one of the other variants.

  • Update 2025-11-01 (PST): - Chomper: A variant designed to release large payloads into space, regardless of the specific mechanism used to release the payload. (AI summary of creator comment)

  • Update 2025-08-28 (PST) (AI summary of creator comment): Mars Cargo clarification

    • Must be a feature-complete vehicle: test articles with missing or deliberately non-functional major systems do not count.

    • All necessary systems must be present on the flown vehicle (not split across multiple flights). The hardware does not need to function successfully in flight.

    • Counts if SpaceX states the launched vehicle could feasibly deliver cargo to Mars (minimum-viable approaches like landing without legs are acceptable if so stated).

    • Must launch before Jan 1, 2028.

Market context
Get
แน€1,000
to start trading!
Sort by:
reposted

I think this one's priced far too optimistically right now

reposted

Bumping this one. It's been quiet for a while, and I think the probabilities have gotten a bit miscalibrated. (As the author, I'm not going to trade in it though).

@DanHomerick Mars Cargo definition
Design must allow it to deploy payload to Mars surface.

If any 2026 launch attempt does not intend to land so that does not have payload unloading.

Suppose there are test versions that fly in 2027 as tests prior to 2028/9 launch & TMI and are designed to have unloading on mars but this equipment is not on the 2027 test flight but is designed and/or exists on Earth..

Does this qualify as Design allows unloading (as per wording)?
Or does all the necessary equipment have to actually "fly" before 2028? Also, does it all have to fly on one ship or would different things tested on different test flights before 2028 be ok?

@ChristopherRandles I'm going to lean on what I wrote for the HLS variant:

"A test article with major systems that are missing or deliberately non-functional (including life support, airlock, etc) won't count."

For a Mars Cargo variant to have flown, it too needs to be a feature complete specimen, with all the things it needs in order to theoretically deliver cargo to the surface of Mars.

It doesn't have to work as intended. I think they recently mentioned a Mars landing without landing legs -- landing on just the engine bells or skirt. That's very "minimum viable product", but if they think it could feasibly work out, well, they're the experts.

Bottom line is if they launch a vehicle which they say they think would be able to deliver cargo to Mars, and the launch happens before Jan 1, 2028, then that's a YES.

@DanHomerick Can a single launch of a HLS demo ship cause 3 questions to resolve yes i.e. Do HLS, Mars cargo and Mars crew all resolve yes if SpaceX claim it could be fuelled sufficiently to send to Mars and have a chance of surviving?

A couple of possible wrinkles here:

1. Only one ship that might have low chance of successfully landing on Mars - well obviously it wouldn't actually be sent with crew without testing similar ships uncrewed. If you only have one such ship launched and it is assigned to lunar missions then you cannot do that testing.

2. Would regulatory reasons why the ship could not be sent to land on Mars mean Mars cargo and /or Mars crew options could not resolve yes? Regulatory reasons could include planetary protection reasons or human rating regulations until it has satisfied requirements not to be too dangerous to launch with or otherwise send crew.

I see ship has to have a life support system. If it has one but it is turned off because no humans are on boards breathing out CO2. Most things on spacecraft can be remotely turned on but supposing it either cannot be remotely turned on or we cannot get info on whether it could be remotely turned on, does that still count as sufficient or does this fall into the wording of "deliberately non functional"?

@ChristopherRandles I don't think a multi-resolution scenario is plausible. In particular, HLS doesn't need atmospheric entry heat shielding, whereas anything Mars-bound will.

As for crew vs cargo variants of a Mars bound craft, yeah, it might get a little fuzzy if they kit it out with some crew support purely for unmanned testing. I think in practice, I think it'll be pretty clear if and when each variant flies, or whether they end up only having a crew varient that fills both roles.

Realistically, the "by 2028" deadline moots most of it. They're going to having their hands full just getting an HLS launched.

@DanHomerick >"HLS doesn't need atmospheric entry heat shielding, whereas anything Mars-bound will."

Well I did think that: I expected lunar HLS would have MLI insulation to prevent prop boil off and i assumed the smooth white look of HLS renders was this MLI under an aerocover. Mars needs heat shield so hexagonal tiling expected.

However I have become less sure about this. Lunar HLS may need micro-meteoroid protection and there were some white hexagonal tiles being tested that were speculated to be HLS related. So what if lunar HLS ends up with white hexagonal tiles for prop boil off reduction and also for micro-meteoroid protection and it just so happens these tiles might be good enough mars re-entry heating wise to have a shot at a landing?

However you are probably right for a slightly different reason: Mars ship likely needs flaps/canards whatever they should be called to control attitude in re-entry whereas lunar HLS (first ones disposable rather than being capable of landing on Earth) probably won't have them.


HLS kitted out for crew might still be usable for cargo, even if a lot of mass on life support and human accommodation which wouldn't be needed and inefficient to have it on a cargo launch? Could HLS cause chomper to resolve yes, if the only criteria is a large door? (Hatch is probably smaller than pez dispenser door, but bear with me.) Would there have to be something recognisable as a PAF (Payload Attach Fitting) for chomper to resolve? Optimus pushing something out of hatch intended for use on Earth and lunar surface as the deployment mechanism would be a bit mad but wording "regardless of the specific mechanism used to release the payload" might make that interpretation more plausible. However I guess it fails on the "variant designed for" wording. If it is designed for crew then it probably isn't also designed for cargo?

If that is how the interpretation works then I guess HLS having tiles for prop boil off and micro-meteoroid protection is designed for that purpose and it is not designed for mars re-entry even if by some fluke it might be good enough. So yeah I am eventually reaching conclusion that probably no adjustments to question description needed to prevent one ship resolving multiple options. It might however be clearer if the "designed for" wording is more clearly explained and applied to all the variants e.g. add


A ship will be considered to be a particular variant if it is designed for that purpose but not if it is designed for another of the listed variant purposes but just so happens might just about work (probably inefficiently) as another of the listed variants.

In the description I said "payload area" but the more I think about it, the more I'm convinced that a Tanker variant simply won't have a payload area. That is, they will stretch the main LOX and LCH4 tanks to take up the whole rocket, and will have leftover propellent in there upon reaching orbit. They will transfer to another ship via the same hardware used for filing the tanks on the ground.

@DanHomerick Yeah it's unclear at this point but either way I expect the tanker variant to be pretty unambiguously a tanker.

I won't bet on this market (because I'm classy like that), but I will kibitz!

I think ya'll are betting the probabilities too high, in aggregate. Excluding Pez, there are 7 variants, and they're all are over 50% right now. Remember that each has to fly in the next 35 months, so that's a new feature-complete variant introduced every five months from here on out. The longer they wait to introduce variant #2, the tighter the schedule for the rest.

SpaceX moves at a ludicrous speed compared to a lot of the old guard aerospace companies, but that's well into "plaid".

To be clear, I will be resolving "Pez Dispenser" to YES upon their Jan 2025 launch. I'm sure they will do more iteration on their design, but Ship 33 appears to have all major systems in place, making it a representative of the variant.

Makes sense to me! It can dispense things Pez style.

Does a different type of cargo door also count as a chomper? Or does it have to hinge like a jaw?

@Mqrius "Chomper" is just because it needs a name rather than trying to be descriptive. If SpaceX flies a variant designed to release large payloads into space, it will count for the Chomper category.

reposted

I'm pretty happy with this market. It's interesting that we're expecting so many variants at all, regardless of timelines!

@DanHomerick Thank you for making it! I'd been meaning to create a market like this for ages but hadn't gotten around to it ๐Ÿ˜