SpaceX progress toward 25 Starship launches in 2025 (now 2026)

[catdlr]

Here is from a potential commercial user for Starship and his needs:

TBPN@tbpn
"It's extremely difficult to imagine what's about to come down the line with Starship."

"We're really talking about 1000x more tonnage per year that we can get to space. And that's coming in three to five years, which is really not that far away."

"The launch cost might come down by between 10 and 100x. Launch capacity is set to go up by 1000x or more."

"The reason is if you build a new Starship every day for a year, which is a very conservative estimate for what
@SpaceX is trying to do, at the end of the year you have 365 Starships, which have 5x the current payload capacity, and you're launching 10x as frequently."

Starcloud CEO @PhilipJohnst0n on the huge change @elonmusk's Starship represents for how much payload we'll be able to take to orbit.

https://twitter.com/tbpn/status/1982941438200717777


More on this at this thread:
https://forum.nasaspaceflight.com/index.php?topic=61453.msg2621818#msg2621818

[spacenut]

So, what are the bottlenecks?

Raptor production, especially R3?

Superheavy production?

Starship production?

TPS production?

Launch facility construction?

Manufacturing production and facilities?

Otherwise, they should be launching more. 

[DanClemmensen]

So, what are the bottlenecks?

Raptor production, especially R3?

Superheavy production?

Starship production?

TPS production?

Launch facility construction?

Manufacturing production and facilities?

Otherwise, they should be launching more.

This thread is about 2025. The "bottlenecks" were:
   *time required to evaluate test failures
   *time required to modify the next ship to accomidate the lessons learned
   *the major mishap at Massey's and time taken to work around the loss
   *shift to V3 and resulting inability to launch after IFT-11

Your list of bottlenecks may affect 2026 and you should probably take them to a new thread. I do not think any of them affected the 2025 launch rate.

[Cabbage123]

SpaceX launched 5 times out of 25 this year, anyone want to calculate how many launches before 2028 would be required for this percentage?

https://x.com/elonmusk/status/1982442502956212305/

Despite significant growth by our competitors, especially China, SpaceX will probably increase from ~90% of Earth’s total payload to orbit to ~98% in about 24 months due to Starship

Obviously a very rough and ready way of doing it, with lots of assumptions, but going from 90% to 98% is 5 times the amount of payload right?

Approx 130 Falcon 9 flights in 2024, if that was ~90% then you would need 650 F9 flights to get to ~98%.

F9 non-expendable mass to LEO is ~17 tons, v3 Starship could be 200 tons.

650/200*17= ~55 flights per year (assuming that Elon is saying that, from 2028, 98% of new payload will be SS).

So, if you just used Starship for all flights, maybe 55 flights with so, so many assumptions (including not knowing what percentage of the 17 tons was actually used in 2024).

Quite possibly less, as with more capacity you can also have less wasted space.

[DanClemmensen]

SpaceX launched 5 times out of 25 this year, anyone want to calculate how many launches before 2028 would be required for this percentage?

https://x.com/elonmusk/status/1982442502956212305/

Despite significant growth by our competitors, especially China, SpaceX will probably increase from ~90% of Earth’s total payload to orbit to ~98% in about 24 months due to Starship

Obviously a very rough and ready way of doing it, with lots of assumptions, but going from 90% to 98% is 5 times the amount of payload right?

Sorry, but no. You first need to estimate the world's total payload in the target year, and then estimate how much of it is non-Spacex. Neither of these is simple. For example, non-SpaceX might remain fixed, and SpaceX add all new payload:
     in 2025 non-SpaceX=1, SpaceX=9.        (90%)
     in 2028 non-SpaceX=1, SpaceX=49.      (98%)

Or, SpaceX might take lots of business away from other launchers:
     in 2025 non-SpaceX=10, SpaceX=90    (90%)
     in 2028 non-SpaceX=2, SpaceX= 98    (98%)

[meekGee]

So, what are the bottlenecks?

Raptor production, especially R3?

Superheavy production?

Starship production?

TPS production?

Launch facility construction?

Manufacturing production and facilities?

Otherwise, they should be launching more.

This thread is about 2025. The "bottlenecks" were:
   *time required to evaluate test failures
   *time required to modify the next ship to accomidate the lessons learned
   *the major mishap at Massey's and time taken to work around the loss
   *shift to V3 and resulting inability to launch after IFT-11

Your list of bottlenecks may affect 2026 and you should probably take them to a new thread. I do not think any of them affected the 2025 launch rate.

Exactly.  You can't extrapolate number without understanding trends. I mean you can, and people do, but it's inherently flawed...   Extrapolation works when the underlying processes remain relevant.

Ship lost a year due to life sucking. but that doesn't mean you take the "wish/reality" ratio and apply it from now till forever.

People did that with F9 during the "life sucking" couple of years, and how accurate did that turn out to be?

[Cabbage123]

Sorry, but no. You first need to estimate the world's total payload in the target year, and then estimate how much of it is non-Spacex. Neither of these is simple. For example, non-SpaceX might remain fixed, and SpaceX add all new payload:
     in 2025 non-SpaceX=1, SpaceX=9.        (90%)
     in 2028 non-SpaceX=1, SpaceX=49.      (98%)

Or, SpaceX might take lots of business away from other launchers:
     in 2025 non-SpaceX=10, SpaceX=90    (90%)
     in 2028 non-SpaceX=2, SpaceX= 98    (98%)

Sure Dan. Two assumptions (none of which I listed) is that we are looking at the upper range of Starships needed to reach 98% and that no other launcher is going to gain significant extra payload.

The lower range would be, as you say, if SpaceX cannibalises other launchers (which, of course, is very likely), but the upper end would be if it is all new payload (including Starlink).

In that case you would need to increase 5 times.

Of course, is other launchers did gain more payload, then you need to do more to reach 98%. I don't know what those figures are and so just extrapolated 2024 (and assumed Elon was claiming 90% for that year).

[Vultur]

So, what are the bottlenecks?

Raptor production, especially R3?

Superheavy production?

Starship production?

TPS production?

Launch facility construction?

Manufacturing production and facilities?

Otherwise, they should be launching more.

This thread is about 2025. The "bottlenecks" were:
   *time required to evaluate test failures
   *time required to modify the next ship to accomidate the lessons learned
   *the major mishap at Massey's and time taken to work around the loss
   *shift to V3 and resulting inability to launch after IFT-11

Your list of bottlenecks may affect 2026 and you should probably take them to a new thread. I do not think any of them affected the 2025 launch rate.

Yeah. They basically lost ~half a year due to failures, and the downtime before V3 will eat up more.

at this point it's about how smoothly the first couple flights of v3 go.

[DanClemmensen]

Sorry, but no. You first need to estimate the world's total payload in the target year, and then estimate how much of it is non-Spacex. Neither of these is simple. For example, non-SpaceX might remain fixed, and SpaceX add all new payload:
     in 2025 non-SpaceX=1, SpaceX=9.        (90%)
     in 2028 non-SpaceX=1, SpaceX=49.      (98%)

Or, SpaceX might take lots of business away from other launchers:
     in 2025 non-SpaceX=10, SpaceX=90    (90%)
     in 2028 non-SpaceX=2, SpaceX= 98    (98%)

Sure Dan. Two assumptions (none of which I listed) is that we are looking at the upper range of Starships needed to reach 98% and that no other launcher is going to gain significant extra payload.

The lower range would be, as you say, if SpaceX cannibalises other launchers (which, of course, is very likely), but the upper end would be if it is all new payload (including Starlink).

In that case you would need to increase 5 times.

Of course, is other launchers did gain more payload, then you need to do more to reach 98%. I don't know what those figures are and so just extrapolated 2024 (and assumed Elon was claiming 90% for that year).

Or alternatively, the Chinese might double their payload mass or even double the non-SpaceX payload mass, in which case SpaceX would need a factor of about ten increase.

We also have this weird problem if how to count the mass of a Ship that will stay in space and operate there for months or years. Is is a second stage (zero payload) or is it a spacecraft that has been delivered to orbit? Would that be the dry mass plus whatever fuel was left after it reached orbit? We would certainly count the whole mass of a Blue Moon MK II, so what is the equivalent for Starship HLS?

[thespacecow]

Of course, is other launchers did gain more payload, then you need to do more to reach 98%. I don't know what those figures are and so just extrapolated 2024 (and assumed Elon was claiming 90% for that year).

Elon is claiming 90% for 2025, not 2024.

And they still have some way to go, at least in Q2 they didn't hit 90%, it's about 86%: https://x.com/FutureJurvetson/status/1969855376129659252

[AmigaClone]

Of course, is other launchers did gain more payload, then you need to do more to reach 98%. I don't know what those figures are and so just extrapolated 2024 (and assumed Elon was claiming 90% for that year).

Elon is claiming 90% for 2025, not 2024.

And they still have some way to go, at least in Q2 they didn't hit 90%, it's about 86%: https://x.com/FutureJurvetson/status/1969855376129659252

Based on the Brice Reports SpaceX is responsible for ~86% of the known mass to orbit by launch provider in the first half of 2025.

https://brycetech.com/reports/report-documents/bryce-briefing-2025-Q1/
https://brycetech.com/reports/report-documents/bryce-briefing-2025-Q2/

Some launches don't include the payload mass.

[thespacecow]

Not precisely on topic but this thread is the closest I can find for discussing launch cadence ramping.

What would a “simplified” Starship plan for the Moon actually look like?
The problem is that it may be difficult to find options that both NASA and SpaceX like.

Eric Berger – Nov 13, 2025 7:09 AM |  68

From the article:

There is a third downside, and this is perhaps the most important one. An “expendable” Starship plan would be anathema to the leadership of SpaceX, including founder Elon Musk. Officials there do not believe the space industry has fully digested how Starship will transform the launch industry.

“You don’t yet understand how many Starship launches will happen,” a senior SpaceX source told Ars.

The company is aiming to launch 1 million tons of payload to orbit per year, the majority of which will be propellant. SpaceX simply believes that once it locks in on Starship operations, launching a dozen or many more rockets per month won’t be a big deal. So why waste time on expendable rockets? That era is over.

[thespacecow]

Obviously Elon Time, but indicates the intention to ramp up cadence to extremely high numbers, fits well with what the SpaceX source told Berger: https://x.com/elonmusk/status/2011676996208640336

In about 3 years or so, Starship will launch more than once per hour

https://twitter.com/elonmusk/status/2011724029435265262

Might take 4 years, but not more than that

[meekGee]

I forgot this thread existed even.

Much has changed since December 2024.
On the down side, 2025 became "the year of the COPV". Several losses, some in the air, some on the ground. Schedule wise, 2025 (and v2) were almost a net-zero.  In somewhat of a recovery, in the last two flights of v2, everything went well, finally.  But still, afterwards, B18.

On the up side, we're seeing the true scale of the Starship program take shape: not just the single rocket, but the launch and manufacturing infrastructure, enough to manufacture on the order of 1 ship per day from each of two factories, and fly 10 per day per tower from each of at least 4 towers in the near term.

Musk's quote about 1 ship per hour in a few years aligns with this.

I think it makes sense to carry this thread over to 2026 rather than start a new one - perspective is everything.

[ZachS09]

I replied to Elon’s “more than once an hour in three years or so”, questioning how the FAA and environmentalists will react. Especially with an average of maybe 50 sonic booms per day, assuming booster catches are done during every launch.

[meekGee]

I replied to Elon’s “more than once an hour in three years or so”, questioning how the FAA and environmentalists will react. Especially with an average of maybe 50 sonic booms per day, assuming booster catches are done during every launch.

Yup, many questions.

On booms, best that I could tell, we're basing our expectations on Shuttle returns.  I was trying to ascertain the effects of the different flight profile (much lower lift) of Starship on that.  It'll fly further higher, and drop more quickly - but to what effect, I don't know.

Regarding booster, just from my experience, "out of the blue" booms are infinitely more disturbing than pre-known ones, and Booster essentially gives you a roaring heads up that a boom is coming. At least during the daytime, this will factor in.

But yes - it'll sure be an issue. If you think Starlink was something, you ain't seen nothing.

[sdsds]

Coincidentally there's a nice sequence: 1 + 3 + 5 + 7 + 9 = 25

SpaceX could launch 1 in 1Q26, 3 in 2Q26, 5 in 3Q26, 7 in 4Q26 and then 9 in the special Elon universe 5Q26.

Edit to add a more serious attempt, looking at two-month intervals:
(1, 2, 3, 4, 5, 10) → 25 ✓
(1, 2, 3, 4, 6, 9) → 25 ✓
(1, 2, 3, 4, 7, 8 ) → 25 ✓
(1, 2, 3, 5, 6, 8 ) → 25 ✓
(1, 2, 4, 5, 6, 7) → 25 ✓

That's it, assuming that in each two-month interval they launch more than in the previous, i.e. no 'backsliding.'

[crandles57]

Is 25 even remotely possible for 2026?

Flight 12 suborbital in April, if that goes smoothly then they can start ramping up production rate but I don't think the production line will be full swing while they wait for first v3 flight results. Also not sure if they will attempt recovery of booster 19 so may have to also build booster 20. 6 weeks to orbital flight 13 at end of May 2026 might be rather optimistic unless they are able to start getting on with stacking ship 40 and booster 20 soon.

Even if they recover ship 40 from flight 13 there are likely lessons to learn before reuse or at least some time to refurbish before reflight can be attempted. If ships 41 and 42 are soon ready after ship 40's flight 13 then they will have been built without flight 40 result knowledge and either cannot be reflown or need changes retrofitted. So more likely fair bit of delay before ships 41 and 42 are ready for their first launch so maybe end of July for the first 4 flights and very optimistically ships 41 and 42 are pez dispenser versions but with fuel transfer test equipment and these two ships are available for reuse.

As additional ships become available from being built and from reuse and if pad and booster turnaround allow, maybe you can maybe optimistically imagine getting to
2 launches in August,
3 launches in September
4 launches in October
5 launches in November
6 launches in December

That gives a total of 24 launches for 2026 with everything running perfectly smoothly. I am not sure the pad can be turned around fast enough for 6 launches in December and even if that pad turnaround is possible, it may well require a few boosters which probably slows down the rate at which new ships are being added to the fleet making that launch rate ramp hugely optimistic.

So 25 launches in 2026 seems crazily optimistic to me, I don't really see how it could be possible. Much more likely there is still an anomaly or 2 with v3 and even 10 launches in 2026 might not be achieved.

[Vultur]

Yeah, I think 2026 will still be slow, especially if Flight 12 really does wait until April. Rapid ramp up may occur later in the year if early flights go well, but probably not soon enough to get a high total for the year.

2027 on the other hand ...

Once/hour by 2029-2030 seems ridiculous to me though. I don't think either the infrastructure or the demand will ramp up nearly that fast.

[xvel]

There just isn't production capability for more than a stack/2 months and there will be no increase this year, first gigabay will probably just start working at the end of this year and spit out first vehicle next year, 10 flights with reusing boosters is absolute maximum that is physically possible, but there are a lot of issues on top of that, the fact that S40 didn't start stacking yet is evidence that they aren't sure about V3 design and so they are delaying production. I hope for 6 flights but each day even this looks less and less probable.