Stocks for Thoughts · The SpaceX Trade

Five futures, one ticker.

The same company prices out five different ways by 2030. Four of them need no new physics — cheaper launch, a durable Starlink, a GPU‑leasing arms dealer, a defence‑and‑connectivity conglomerate. Only the fifth needs the thermos to work. Today's ~$1.5T sits between the Conglomerate and the leap.

valuation range · needs no new physics needs orbital cooling to work / physics flag
Scenario · 2030 revenue Implied valuation  → Range
1The Unwind
Curves never meet. A good broadband utility carrying an expensive habit.
no new physicsrev ~$30–40B
$0.3–0.5T
2The Utility
Cheap launch, a durable Starlink. Repriced as infrastructure.
no new physicsrev ~$50–65B
$0.6–0.8T
3The Landlord
Sells picks, not models. The only branch already generating cash.
no new physicsrev ~$80–100B
$0.9–1.2T
4The Conglomerate
All three engines run at once. Grows into itself for half a decade.
no new physicsrev ~$90–120B
$1.3–1.6T
5The Orbital Hyperscaler
The curves cross. Compute in orbit is permanently cheaper than a grid.
needs the thermosrev ~$200–350B
$3.5–5T
today · ~$1.5T
Company value →
$0.3T $0.5T $1T $2T $5T
Read it as a ladder. The four grounded futures top out around $1.6T — roughly where the stock trades today. Everything above that line is the market pricing in some chance of the fifth: the orbital branch, the only one that needs cooling physics nobody has demonstrated at scale. Whether today's price is a moonshot or a mispricing is the size of the bet on that top rung.
Scale: horizontal axis is company value on a log scale, $0.3T→$5T; bars span each scenario's illustrative valuation range, revenue is 2030E. Sources: SpaceX Form S‑1 / 424B4; Goldman Sachs 2030 revenue ~$474B (AI $322B / Starlink $144B / launch $8.3B); Morgan Stanley ~$330B; company value ~$1.5T reflects late‑July 2026 trading (peak near $2T at June IPO). Illustrative scenarios, not forecasts. Not investment advice.