Ben Okonkwo: So I've been waiting to see how you're going to frame this one, because I have a feeling you're going to lead with the win and I'm going to have to be the one who brings up the Gulf of Mexico.
Eleanor Crane: You know me. But — no, actually, I want to lead with the take that's going to annoy you, and then I want you to take it apart. Ready?
Eleanor Crane: July 24th, Starbase, Boca Chica. IFT-13 lifts off. Twenty Starlink V3 satellites — not test mass, not dummy payload — twenty actual next-generation Starlink satellites get deployed from a Starship for the first time in the program's history. The upper stage comes down in the Indian Ocean, west of Australia, floating, transmitting. Elon Musk is on X: 'Starship is intact, floating in the ocean and transmitting telemetry.' And the market, which has been watching since the SpaceX IPO closed in June, is cheering. The booster crashed into the Gulf again. And I'm saying: that's a footnote.
Ben Okonkwo: No, I don't buy that.
Eleanor Crane: Tell me why — and specifically, because I think the narrative pull here is real. Twenty satellites in orbit is not nothing.
Ben Okonkwo: The upper stage wins are real — Raptor relit, softest splashdown yet, actual payload delivered. I'll grant all of that. But the Super Heavy booster is not a sidebar. It's the piece that's supposed to land and reflly within days. It has now failed the landing burn in exactly the same way on Flight 12 and Flight 13. That's a pattern, not noise. And the first time that pattern shows up, SpaceX is a public company.
Eleanor Crane: Which is the thread I want to pull today — because what changes when the same crash happens under a stock ticker.
Ben Okonkwo: Think of it this way — and this is the thing that settled it for me. Imagine an airline that proves, every single flight, that the fuselage lands safely. Passengers walk off fine. But the engines are destroyed on touchdown. Every time. You can't run a schedule on that. That's the Super Heavy situation right now.
Eleanor Crane: The booster is the engine.
Ben Okonkwo: The booster is the business model. Rapid reusability — the whole economic argument for Starship — only works if you recover it and fly it again within days. No recovered booster means that loop is broken. And what we have now is two consecutive flights, six weeks apart, with engine-ignition failures during the landing burn. Not similar failures. Identical.
Eleanor Crane: And it wasn't just in-flight. There was the July 16th pad abort — wasn't that also a booster propulsion issue?
Ben Okonkwo: Frozen moisture in the oxidizer turbopumps. Six Super Heavy booster engines. The problem showed up before the launch even happened, and then showed up again at landing. That's — I mean, that's not a random bad day. That's a propulsion system that is sensitive in ways SpaceX hasn't fully characterized yet.
Eleanor Crane: Which means — wait, does that actually tell us the root cause is the same across both events? Or are these two separate failure modes that happen to involve the same engines?
Ben Okonkwo: That's exactly the thing the data doesn't answer yet. What it does tell us is that the Super Heavy booster has a pattern of propulsion sensitivity — cold, dynamic, relight conditions — and SpaceX is now a public company trying to explain a repeated hard splashdown in the Gulf of Mexico to shareholders who priced in 2028 lunar landings.
Eleanor Crane: And NASA is watching that same timeline. Artemis.
Ben Okonkwo: Right — but the part that doesn't fit the reassuring story is this: you can't land astronauts on the moon with a vehicle whose first stage keeps destroying itself on touchdown. The upper stage success is real. The booster failure is also real. And only one of those two things threatens the entire reusability economics of the program.
Eleanor Crane: But hold on — because I don't want to let the upper stage data disappear into that framing. A single Raptor relit in space. In vacuum. That's not a footnote, that's propulsion control in an environment where you get one shot.
Ben Okonkwo: No, that's real. Raptor relight in space, Starship V3 second flight, tiles surviving elevated dynamic pressure — those are not spin. The drone inspection gave engineers the best post-flight heat-shield data they have ever had. One analyst called it the single biggest remaining reusability risk, now addressed.
Eleanor Crane: And I want to sit in that for a second — because picture a payload engineer. Her team just watched twenty Starlink V3 units power up from orbit. First time in this program's history a Starship carried functional satellites. That moment is new. That's Starship crossing from test article into operational candidate.
Ben Okonkwo: Which matters enormously for the tower catch — the next stated milestone is catching the upper stage with the launch tower arm, and you can only attempt that if you trust the vehicle's intact arrival. This flight earned that trust.
Eleanor Crane: So the tile data isn't just about this flight. It's the foundation for that catch attempt.
Ben Okonkwo: Exactly — and, I mean, the analyst claim is load-bearing in a specific way. Tiles surviving elevated dynamic pressure means the thermal protection system held under conditions harder than previous flights. That's the actual datapoint. Not solved forever, but solved for this regime, and the drone inspection confirmed it structurally.
Eleanor Crane: Which is — well, that's a real kernel. The upper stage partial win is genuine.
Ben Okonkwo: It is. And the part that makes all of this more complicated — which we haven't gotten to yet — is what the booster problem actually does to NASA's 2028 Artemis lunar landing, because the math there involves orbital propellant transfer and serial tanker launches that have never been tested, and none of it works without a rapidly reusable booster.
Eleanor Crane: And that's a very different kind of deadline than a stock price.
Ben Okonkwo: And that deadline is specific in a way that's easy to lose — NASA's Artemis program has a crewed lunar surface landing targeted for 2028. That's not a vague aspiration. And the Starship Human Landing System gets astronauts from lunar orbit to the surface. But before any of that, you need orbital propellant transfer. In-orbit. Multiple tanker launches filling a depot in space. That has never been attempted on any IFT mission. Not once.
Eleanor Crane: Wait — how many tanker launches are we talking?
Ben Okonkwo: Multiple rapid launches in close succession — and each one requires a booster that lands, gets refurbished fast, and flies again. That's the whole point. If the Super Heavy keeps making hard Gulf splashdowns, that serial cadence is physically impossible. You can't stack tanker missions on a booster that destroys itself.
Eleanor Crane: So the 2028 date isn't threatened by one bad flight. It's threatened by — I mean, it's more structural than that. It's threatened by the absence of a demonstrated capability that no flight has even tried yet.
Ben Okonkwo: That's the precise version of it, yeah.
Eleanor Crane: Which means the media framing split — some outlets leading with booster failure, SpaceX emphasizing mission success — that split isn't really about spin. It's about which audience you're writing for. Shareholders can absorb an upper-stage win. NASA cannot absorb an untested propellant transfer system eighteen months from a crewed landing.
Ben Okonkwo: And post-IPO, SpaceX has to talk to both of those audiences simultaneously with the same flight data. That's — actually, that's a genuinely new constraint on how they communicate risk.
Eleanor Crane: So what's the defensible version of Flight 13?
Ben Okonkwo: Inflection point — but conditional. The upper stage result is real, the tile data is real, the Starlink V3 deployment is real. If the booster problem gets resolved in the next two flights, Flight 13 is the moment the program turned. If it doesn't, the intact upper stage floating in the Indian Ocean is genuine but irrelevant to the critical path — because no recovered booster means no tanker cadence, no propellant depot, no 2028 Artemis landing. The astronauts don't go.
Eleanor Crane: Fine. The booster is the story. I just wish the Super Heavy booster had read the same script as the upper stage.
Ben Okonkwo: That's — honestly, that's the cleanest version of it. Flight 13 proved Starship can carry real Starlink V3 satellites, come home intact, float in the Indian Ocean transmitting telemetry. All of that is true. It also proved, for the second consecutive time, that the Super Heavy booster fails in exactly the same way it failed six weeks ago. NASA has a 2028 Artemis deadline. The booster doesn't care about deadlines.
Eleanor Crane: We started today with Elon Musk's post about Starship floating and transmitting — and I read that as a win. I think I read it wrong. Or, I mean — not wrong exactly. Just incomplete. He didn't mention the Gulf.
