SpaceX's First 100 Days Are in the Books. Here's the Report Card.

SpaceX shares closed at $154.72 on Sept. 22, roughly flat versus their $150 June 12 IPO opening price, after falling more than 50% peak-to-trough in July and recovering. Q2 revenue rose 92% to $7.8B and adjusted EBITDA nearly tripled to $3.5B, driven by AI compute-rental agreements with Anthropic and Google and 66% connectivity revenue growth as Starlink subscribers doubled. However, the company reported a $541M Q2 GAAP loss and remains expensive at a roughly $2T market capitalization despite a lower valuation than at its approximately 100x sales IPO multiple. Two lockup tranches passed without sustained selling pressure, but the article views SpaceX as less attractive than more reasonably valued large-cap AI beneficiaries.

Analysis

The key underwriting question is not growth rate but revenue quality. Compute leasing can inflate reported growth and EBITDA while carrying materially lower durability, customer concentration, renewal visibility and capital intensity than recurring connectivity or launch-related revenue. If this mix persists, SPCX should ultimately trade closer to GPU-cloud comparables such as CRWV/NBIS on the incremental revenue stream, rather than receive a blended premium multiple associated with a scarce strategic platform.

The first post-IPO liquidity events are a weak read-through for the remaining supply overhang: early unlocks often involve employees and investors with lower urgency to monetize, while later tranches can coincide with hedging, tax sales and portfolio rebalancing. Moreover, greater float does not itself create mandatory QQQ demand; Nasdaq-100 weighting depends on index eligibility, free-float-adjusted market capitalization and reconstitution mechanics. The next quarterly filing is the nearer-term catalyst: investors need segment-level bookings, contract duration, capex, depreciation and customer concentration to determine whether EBITDA conversion is repeatable.

Consensus may be too focused on the absence of immediate selling pressure and too little on the mismatch between a premium equity multiple and GAAP cash economics. A sustained re-rating requires evidence that the core recurring businesses—not opportunistic infrastructure utilization—can fund the capex burden while expanding free cash flow. Conversely, validated long-duration contracts with prepayments and limited incremental capex would make the current skepticism too conservative and pressure shorts quickly.

Key Decisions for Investors

• Maintain SPCX as a watch/underweight rather than chase post-lockup stability; reassess after the next results release only if management discloses contract tenor, remaining performance obligations, customer concentration and segment capex. A short thesis is falsified by durable free-cash-flow conversion and multi-year contracted backlog rather than another headline revenue beat.
• Express relative valuation risk via long META / short SPCX over the next 1-3 months, sized beta-neutral. META offers more established cash conversion and AI monetization; target a 10-15% relative move, with a stop if SPCX delivers disclosed, long-duration compute contracts plus upward full-year free-cash-flow guidance.
• Avoid shorting CRWV or NBIS solely as read-throughs: incremental third-party compute supply can pressure pricing over 6-18 months, but a large new buyer of GPUs, networking and power capacity may initially support the broader AI infrastructure complex. Monitor GPU utilization, lease pricing and power commitments before initiating any pair trade.
• For event-driven mandates, consider a defined-risk SPCX put spread dated just beyond the next earnings report only if implied volatility remains below the prior post-results move. The catalyst is disclosure that leasing revenue is short-term or capex-heavy; do not initiate without options liquidity and borrow data.