SpaceX (SPCX)

Executive Summary & Action Plan

Verdict

SELL (re-affirmed on re-mark) — we stress-tested yesterday's mark against three new axes and all three point the same way. The fair-value methods held at $45.60: the 24-hour filing sweep found nothing FV-moving, and third-party space/AI TAM evidence judges our 17% credit to the $1T target "about right." What moved is the distribution: a hike-tilted Fed with no rate-cut put and five months of staggered unlock supply cap the bull, while observable AI-credit stress and Alphabet's saleable stake fatten the bear — weights go 20/55/25 → 15/55/30, weighted fair value $47.9 → $43.60. Spot sits 10.6% above even the bull case.

Price $108.27 | Market cap $1.43T | Target $45.60 | Upside −57.9% IWANNAVY Fair Value (log-median of 3 methods, held on re-mark; probability-weighted $43.60 at 15/55/30) · price as of 2026-08-05 close, verified on two sources exactly · fully-diluted 13.746B shares · Street $223–$234 (n=31–35, reference, ~2.1x spot) — street extrapolates the $100B run-rate guide; no third-party 2030 estimate comes within half of the $1T target

Abstract

Today ~911.5M SpaceX shares (~$98.7B) become sellable — a 2.4x float expansion into a tape that already front-ran the event with a −13.6%, 208M-share capitulation on August 5 and a +1.6% after-hours bid. We re-marked the model on three new axes. Flows: the unlock shapes path, not value — the modal analogue is Facebook's November 2012 "sell the rumor, cover the news" relief, the tail is Rivian 2022 if Alphabet (~$80B short-term-saleable of its $94.1B stake) or a top VC announces a block. Macro: a hawkish-hold Fed with a 62%-priced September hike, 10Y at 4.62%, and richest-decile credit spreads removes the re-rate put; AI-financing stress is now observable. Industry: the space economy compounds ($686B in 2025, +12%), but the highest sell-side 2030 revenue estimate is ~$470B versus the $1T target — our 17% credit stands. Methods held at $45.60; weights tilt to 15/55/30, weighted $43.60. SELL re-affirmed; the relief bounce, if it comes, is an exit.

Forecast Path

Weights re-derived at 15/55/30 (non-default): bull −5pt on the removed rate put and drip supply, bear +5pt on AI-credit stress plus the Alphabet flow tail, base held on print quality and raised hyperscaler capex guides ($725–785B, +77% YoY). Reversion triggers back to 20/55/25: September-hike odds fading below ~20% (soft 8/7 payrolls, WTI < $70) or the 10Y breaking below ~4.40%.

Action Plan

Sell into strength — the FB-2012-style relief bounce is the modal near-term tape and it is an exit, not a signal; re-entry only in deep fair-value territory, and the sell call itself covers on a sustained re-rate above $150.

• Risk/Reward on the exit call 1.5:1 (−57.9% to base FV avoided vs +38.5% to the $150 invalidation) | Prob-weighted 12M return −59.7% | Confidence: Medium-High on the mark (three methods held under a three-axis stress test), Low on the two-week tape (squeeze window)

---

The Unlock & Flow Map

Eligibility is not selling — the calendar below is the supply ceiling, and holder behavior is the swing variable. As of this morning's pre-open sweep, SEC EDGAR shows zero Form 144s on file and no insider Form 4 since 6/17; the first hard evidence of who accumulated (and who is leaving) arrives with the ~8/14 (E) 13F window and the first post-unlock Form 144 prints.

Three flow facts frame the day. First, the short is optically crowded but mechanically not: 165M shares short is 25.6% of the old 646M float but only 1.44 days to cover, and today's float expansion dilutes it to ~10.6% while flooding the borrow market — full covering would absorb just 18% of the tranche, so a structural squeeze requires holders to sit tight (the ARM-2024 pattern), while cheap borrow makes new shorts easier from here. Second, index demand is lumpy and late: Nasdaq pre-booked part of the float expansion at the July NDX fast-entry (~$300B effective index value), so today triggers no forced buying; the passive offsets arrive at the ~9/18 (E) NDX float true-up, a Sep-or-Dec Russell entry, and the ~12/18 (E) reconstitution — S&P 500 is a mid-2027 story at the earliest (four quarters of GAAP profit required). Third, the analogues bracket the outcome: Facebook Nov-2012 (telegraphed unlock into a crushed, shorted tape → relief rally) is the modal case after yesterday's capitulation; Airbnb/DoorDash staggers say drip supply caps rallies into October; Rivian 2022 is the tail — one announced Alphabet or VC block trade converts relief into breakdown. Net: supply-dominated for 1–3 months, cleaner after December; all of it path, none of it value.

Macro Overlay

The regime is a hawkish hold: the Fed left 3.50–3.75% on 7/29 with three dissents for a hike, the market prices a 62% September hike, and the 30Y at 5.20% shows where long-duration discounting already trades. There is no rate-cut put under a ~165x-TTM-EV/EBITDA asset in this tape. The AI capex cycle — SPCX's revenue engine and its funding risk — is being revised up ($725–785B Big-4 guides, +77% YoY), which protects the base revenue path, but its funding quality is degrading: 32% of hyperscaler capex is now debt-funded (vs 9% in FY24), $182B of AI bonds have printed YTD, and CoreWeave 5-year CDS at ~855bp implies ~50% default odds in the comp set. WACC holds at 10.5% (rf updated 4.64% → 4.62%, inside the rounding grain); the macro risk is expressed in the weights and a published $38.80 stress leg (beta 1.45 / WACC 11.0%), not in a silent ERP bump.

Space Industry Outlook

The industry thesis is intact and is not the debate. The space economy reached $686B in 2025 (+12% YoY) and tracks toward ~$1T by the early 2030s; US defense space is the fastest wallet (Space Force FY27 request just over $71B — more than double FY26 — inside a $185B Golden Dome program), and roughly $10–13B (E) of 2026 award announcements corroborate SPCX's $47.5B audited backlog. SPCX's operational dominance is near-absolute: >80% of global upmass in 2025, Starship's first revenue-bearing flight in July, and a launch competitor set that is stalled (Blue Origin's post-anomaly cadence, Neutron's small scale). Starlink has outrun the research TAMs — its $17.2B annualized connectivity run-rate already exceeds several published "2030 satellite internet market" forecasts, meaning Starlink is the market — while Amazon Leo sits far behind its FCC milestones and the D2C/terrestrial push (650+ satellites live, T-Mobile commercial, EchoStar spectrum, femtocell buildout) gives a real, if contested, path at the ~$600B/yr US carrier wallet against AST SpaceMobile's 45–60 satellites by early 2027.

The valuation debate is what the equity pays for that position. The $1T-by-2030 target needs a ~120% five-year revenue CAGR and would require AI compute revenue of $700B+ — 2.6x Gartner's entire 2030 AI-cloud TAM. No third party underwrites it: post-print sell-side 2030 estimates top out at ~$470B (GS; MS ~$330B), and at $108.27 the market still pays for ~$323B (E) of 2030 revenue — roughly the most bullish bank's number. Our 17% credit ($168B 2030E) is judged about right with mild upside skew; an upgrade to 18–20% credit (fair value ~$50–53 (E)) is gated on Q3 evidence — October cloud-ramp conversion and ≥3 GW of compute exiting 2026. Own the industry thesis, not this price.

The Print, One Day Later

Full statement analysis is in the 2026-08-05 tearsheet; the frame that matters for the re-mark is unchanged — a genuine P&L inflection funded by a capex supercycle. Revenue $7.81B (+92% YoY), adjusted EBITDA margin 45.3%, AI segment EBITDA-positive for the first time; capex $18.4B with two more quarters guided at that level, FCF −$16.0B against $100B of cash and +$60.6B net cash.

IWANNAVY Fair Value — Re-marked

Methodology continuity was enforced: same three methods, same 13.746B FD share base, same log-median aggregation, and every change (or hold) traced to dated evidence — the audit found no anchoring and no hidden rebuild. Lockup flows were deliberately kept out of intrinsic value: the FD base already counts every locked share, so the unlock changes tradability, not per-share claims; the one channel through which flows could touch value — forcing a dilutive raise — is not triggered ($100B cash, capex funded through Q4'26, zero raise evidence on file).

• Thesis breaker (for the SELL): a Starship tower catch (~8/20 (E)) plus Q3 evidence the December $100B run-rate is real, with capex held and the October cloud ramp converting — that combination re-opens the bull path and lifts the mark toward $50–53 (E); a sustained re-rate above $150 covers the call outright.

---

Catalysts & Risks

---

References

• SpaceX Q2'26 earnings release (8-K EX-99.1, SEC EDGAR, 2026-08-04) · Form 10-Q, period ended 2026-06-30 (SEC EDGAR, 2026-08-04) · SEC EDGAR full-text Form 144/Form 4 sweep (2026-08-06 pre-open, nil)
• Forbes — nearly 1B SpaceX shares unlock (2026-08-04) · Bloomberg — the Great Unlocking, $116B (2026-07-21) · Motley Fool — the 911.5M-share unlock timeline (2026-08-05) · CNBC — earnings date triggered the unlock (2026-07-21)
• Alphabet Form 10-Q (2026-07-23) — $94.1B / 6% SPCX stake, ~$80B short-term-restricted · Space Foundation, The Space Report 2026 (2026-07) — $686B 2025 space economy · McKinsey & Co. × WEF, Space: The $1.8 Trillion Opportunity (2024-04) · BryceTech launch statistics (2026-04) · CME FedWatch (2026-08-04) · FactSet hyperscaler capex funding note (2026-08-01)
• Companion house reports: SPCX first-print underwrite (2026-08-05) · SPCX pre-print refresh (2026-07-25)