Source review: B1076 34-reuse milestone (March 2026)
Summary
| Verdict | Count |
|---|---|
| Consistent | 1 |
| Novel supporting | 1 |
| Merits investigation | 1 |
| Different conclusion | 0 |
| Not relevant | 0 |
Claims reviewed
Claim 1 — Falcon 9 first-stage reused 34 times
Verdict: Novel supporting
Direct operational evidence that the partial-scenario (30 reuses) reuse count is achievable for Falcon 9 first-stage hardware in mature operation. Before this milestone, the 30-reuse target was aspirational. As of March 2026, it's demonstrated.
This is the single most important operational anchor for the q1 calc — it converts the partial scenario from "plausible" to "demonstrated for at least one booster."
Claim 2 — B1076 has launched 22 Starlink batches + 12 dedicated missions
Verdict: Consistent
Supports the model that high-reuse hardware is preferentially assigned to lower-risk missions (Starlink, where SpaceX is its own customer) once it gets old. Implies that SpaceX's effective reuse counts may bifurcate: a few "workhorses" approach the demonstrated 34-flight ceiling, while many other boosters fly 5-15 times before retirement.
The fleet-average reuse count is probably closer to 15-20 than 34. Important constraint for the pessimistic scenario — it suggests the 10-reuse pessimistic is too pessimistic for the typical Starship booster (assuming Starship achieves Falcon 9-comparable hardware longevity).
Claim 3 — Implicit: Falcon 9 reuse evidence transfers to Starship
Verdict: Merits investigation
As Codex noted in the pass-03 audit, Falcon 9 first-stage reuse is not the same as Starship full-system reuse. Starship has:
- A different upper stage (Ship) that itself must be reusable, vs. Falcon 9's expendable upper
- Larger vehicle, more complex thermal protection (heat shield tiles)
- Different propellant (methalox vs RP-1/LOX)
- Different recovery method (catch tower vs droneship)
The operational analogue isn't perfect. Falcon 9's 34-reuse milestone bounds the optimism (Starship probably won't exceed Falcon 9's record in early years) and bounds the pessimism (Starship is designed for reuse from day one whereas Falcon 9 was retrofit). It's a useful guide, not a direct prediction.
Cross-reference
- Validates [pass-02-calc] partial-scenario (30 reuses) as the calibrated central case
- Constrains pessimistic-scenario boundary (10 reuses may be too low even for Falcon 9 fleet mean)
- Falcon-9-to-Starship extrapolation caveat (per Codex audit) carried forward to write pass
What we'd want next
- SpaceX-reported fleet-average reuse count (not in this source)
- Booster failure rate data — how many SpaceX boosters have been lost before reaching 34 reuses? That bounds the reuse expectation actuarially.