SpaceX Earnings Preview (August 4): The First Print Ever, With 911 Million Shares Unlocking 48 Hours Later

TL;DR

• SpaceX reports Q2 2026 results on Tuesday, August 4: the first earnings report in the company's 24-year history. Nobody outside the building has ever modelled this business against a real quarterly disclosure.
• The street expects roughly $6.9 billion of revenue. The known baseline is ugly: a $4.9 billion net loss in 2025 and $4.28 billion lost in Q1 2026, driven by Starship and AI infrastructure spending.
• The print triggers the unlock. Roughly 911.5 million shares, about $100 billion at Friday's close, become eligible to trade on Thursday, August 6, against a float of 555.6 million. The full staircase runs to December 8.
• The stock closed Friday, July 31 at $108.37, a new closing low: about 20% below the $135 IPO price and roughly half the $225 post-listing high, before a single locked share has been sold.
• Weekly options price a ±12.5% move. Backtested against five real lock-up expiries, that weekly put paid once. Trade log below.

When Does SpaceX Report Earnings?

The short answer: Tuesday, August 4, 2026, and the report does two jobs at once. It is the first quarterly disclosure SpaceX has ever made as a public company, and under the staged lock-up agreement the Q2 report is the trigger for the first insider unlock, which lands 48 hours later on Thursday, August 6. A bad print does not just reprice the stock; it releases supply into the reprice. A good print gives eighteen-year holders a better level to sell into. Either way, the two events cannot be separated, which is why this preview treats them as one.

The Board

One column is the quarter. The other is the calendar. Thursday only cares about Tuesday.

What the Street Expects, and What It Cannot Know

Consensus sits near $6.9 billion of revenue for the quarter, against a company that did roughly $11.4 billion of Starlink revenue in all of 2025. Three lines decide whether the number lands:

• Starlink. The last disclosed count was 10.3 million subscribers as of March 31, up 105% from 5.0 million a year earlier, across 164 countries. This is the profitable engine that funds everything else, and the question is not growth, it is whether Starlink's profits are growing fast enough to carry the spending below them.
• The AI compute line. Sell-side estimates see this business jumping from $818 million to about $2.18 billion, the fastest-moving number in the model and the one with the widest error bars. This is also the line the February xAI merger was priced on.
• The burn. Quarterly capital spending is projected near $12.9 billion, with roughly $7.7 billion of it AI-related. Stack that against $6.9 billion of revenue and the shape of the company is plain: SpaceX currently spends nearly two dollars for every dollar it takes in. The losses on record, $4.9 billion for 2025 and $4.28 billion in Q1 alone, are the result.

The valuation math has not changed since our piece on the stock breaking its IPO price: at Friday's $108.37 close the company is worth about $1.42 trillion, roughly 90x trailing sales, and the multiple only works if revenue roughly doubles toward $36 billion. Tuesday is the first hard data point that model has ever received.

The Unlock Is the Second Half of the Print

Everything about Thursday's supply event is in the full lock-up calendar, so here is only what matters for the week: 911.5 million shares become eligible against a 555.6 million share float, a 164% increase in tradeable stock in one day, and the sellers most likely to act are early venture funds sitting on returns near 2,500x for whom the difference between $135 and $108 changes nothing. Musk and his 42% stake stay locked until mid-2027, which removes the worst-case scenario and is the single most underweighted fact in the setup.

The week has already been trading this. The stock fell 3.4% on Friday to that $108.37 closing low with no company news at all, which is what front-running known supply looks like. The July 28 session that dragged SPCX down with the memory complex was the same mechanism with a worse tape.

The Two Ways Tuesday Goes

The bear path is obvious and priced. A wide loss, no Starlink margin disclosure, and a capex number that keeps climbing, followed by scheduled supply into a falling stock. Morgan Stanley, at a $300 target, argues fundamentals are "largely unchanged" and current levels are an attractive entry; 23 of 31 covering analysts say buy. That is precisely the crowding that makes the obvious path dangerous to short at these levels: everyone can see the calendar, and the base rate on lock-up expiries is about -2%, not -12%.

The bull path has one precedent, and it is a good one. Facebook's largest unlock, 777 million shares in November 2012, came after the stock had already been crushed, and it rose 12.6% because the feared supply had been front-run for months. SPCX at half its high, at a new low, two days before its biggest unlock, fits that shape uncomfortably well. If Tuesday's print shows the losses narrowing and Starlink compounding, the unlock gets absorbed and this week is the low. That path requires a good print. It exists.

The Options Angle

The weekly chain prices roughly ±12.5% through Friday, August 7, covering both the print and the unlock. Our full backtest of buying weekly options into five historical unlocks found the put paid once, the call paid once (the biggest unlock of the five), and three times both sides expired worthless. When a 12.5% breakeven meets a -2% base rate, the trade is the premium, not the direction, and we are not paying it in either direction.

• Skip the weekly puts. Right about direction, dead on magnitude is the modal outcome. This is the July lesson in reverse: in memory, realised beat implied all month; here the implied is priced for a crash that history says mostly pre-happens.
• Skip the weekly straddle too, and the strangle variant: 12.5% of spot for a week is a price only a Rivian-2022 outcome beats.
• The trade worth planning is the confirmation trade. If the print is good and the stock absorbs Thursday, the scarcity argument dies and the entry argument begins, with five more 7% tranches to buy through. If the print is bad, the December tail of the calendar gets cheaper for months. Wait for Tuesday night, then decide with the number in hand.

Trade log

#	Stance	Structure	Strikes and expiry	Cost or credit	Spot at writing	Implied move	Breakeven
1	Pass	Long weekly put	$108P, Aug 7	Live chain not sourced; weekly options price ~12.5% of spot	$108.37, July 31 close	±12.5% (week incl. print and unlock)	needs roughly -12.5% vs a -2% unlock base rate
2	Pass	Long weekly straddle	$108 line, Aug 7	~12.5% of spot per the weekly implied	$108.37, July 31 close	±12.5%	needs >12.5% either way
3	Conditional	Post-print shares, first third, only if Q2 shows losses narrowing and Thursday absorbs	No expiry	Market on Aug 7 or later	To be struck vs the Aug 6-7 tape	n/a	Scored against the post-unlock entry if taken

Rows 1 and 2 are passes and get scored like positions: if SPCX moves more than 12.5% this week, both are losing calls and will appear in the next scorecard. Row 3 is a plan, not a position, and only becomes scoreable if its two conditions print.

The One-Line Read

The first earnings report in SpaceX history lands Tuesday and opens the gates on 911.5 million shares Thursday: at $108.37, 20% below the IPO and half the high, the crash the puts are priced for has mostly already happened, so let Tuesday's number pick the side for you and keep the week's premium in your pocket.