SpaceX (SPCX)

Executive Summary & Action Plan

Verdict

SELL — The relief rally met the supply calendar. Four unlock waves through December 8 cap the tape; the 12M path marks $89, 33% below spot.

Price $132.73 | Market cap $1,749.7B | Target $89 | Upside −32.9% 12M market-path target (partial convergence) · intrinsic $42.6–$48.1 (log-median $45.60, held from 8/6) · Street $213.50 (n=16, reference; populations vary $213–234) · price as of 2026-08-20 US intraday

The Take

The relief rally played out exactly as mapped — +29% in two weeks off the August 6 unlock — and then met the calendar. The Day-70 tranche — 319M shares, 7% of the locked pool — went saleable on August 20 exactly as the prospectus scheduled; Musk pushed the Starship ship-catch to "months away"; the stock gave back 5%. None of it moves intrinsic value — the three methods hold at $45.60. What changed is supply mechanics: four unlock waves run through December 8, and Musk's 49% stays locked to mid-2027, so convergence is partial, not full. We stay sellers into strength; the 12M path marks $89.

Action Plan

Sell strength above $138; cover the call only on a sustained re-rate through $150; re-entry interest begins near intrinsic ($46).

• R/R 2.5:1 (exit call: −32.9% to 12M path avoided vs +13.0% to the $150 invalidation; entry gate −0.56 → BUY barred) | Prob-weighted 12M −32.9% | Confidence: Medium | p_thesis_wrong 30%

Forecast Path

Probabilities are held from the 8/6 re-mark at 15/55/30 — its stated reversion triggers (Sept-hike odds <20%, 10Y <4.40%) remain unmet — and are now applied to path bands rather than full FV convergence, since Musk's 6.4B locked shares cap tradable float at ~half the register through the window. Risk's p_thesis_wrong (30%) sits above bull's 15% because thesis failure includes index-absorbed outcomes at the top of the base band.

By the Charts

Two quarters are missing from the public record (pre-IPO); the visible arc is a margin V — operating −41.6% to −1.8% in two prints.

Revenue +92% YoY on volume and mix; the operating loss shrank 93% QoQ while FCF worsened to −$16.8B — profit and cash are moving in opposite directions.

AI is already a third of revenue (+213% QoQ, first EBITDA profit); Starlink is 55% with 12.0M subscribers; launch is 12%.

Three methods sit within $6 of each other; the market pays 2.9x the top of the range.

Fair Value — Two Layers

Layer 1 held: DCF $42.6, target EV/EBITDA $48.1, normalized EV/FCF $45.6 — log-median $45.60, unchanged from August 6 because the filing sweep found no FV-moving evidence and the 10Y drift (+6.6bp) is sub-threshold. Layer 2 softens the old full-convergence assumption to partial convergence: only ~half the register can trade inside the window, so the supply mechanism is real but incomplete — weighted path $89. The layers differ by +95% because a supplied-but-narrative-priced tape can persist for another year; full convergence to $45.60 is now the bear path, not the base. Decomposition (FY27E adj EBITDA $27.3B basis): forward-base growth 0% × EV re-rate −34.1% (61.9x → 40.8x) × payout 0% = EV −34.1% → equity −32.9% on the constant $60.3B net-cash cushion — matches the target gap exactly (0.0pp ✓).

촉매 캘린더

단기 (0~6개월)

The next six months are a supply schedule with a narrative overlay: every rally runs into a dated unlock, and every unlock runs into possible index-flow offsets — the sequencing, not the fundamentals, sets the path.

Risks & Thesis Breakers

• Lockup supply shock — 85% × −25–40%, 0–4M (prospectus-dated: 319M live 8/20 → 319M 9/9 → 328.4M each 9/24, 10/9, 10/24 → 1.3B post-Q3 print → 797.6M on 12/8; ~88% of the register saleable by December).
• Valuation compression — 60% × −30–50%, 3–12M (62x FY27E vs 22x target; one growth-deceleration print removes the only support).
• Alphabet/VC block trade — 40% × −10–20%, event tail (~$80B short-term saleable of the $94.1B stake; Form 144s absent so far).
• Breakers (any voids the SELL): close ≥$150 for 10 sessions, especially through an unlock; 12/8 expiry absorbed with <10% five-day drawdown; Q3 revenue ≥$9.5B (E) with operating profit and FCF loss halved; no Alphabet block within 90 days of full expiry; GAAP-profit streak starts the S&P 500 clock.

Appendix