To The Moon

The uncomfortable framing first: the 7.5% drop is not a market error. It is the correct read of a company that just delivered its best quarter ever and then reminded investors that 93% of its current market cap is priced on businesses that still lose money.

Tesla delivered 480,126 vehicles in Q2 2026, crushing Wall Street consensus of 406,024 by nearly 18%. Up 25% year over year. Up 34% from Q1. Its strongest second quarter ever and its first year-over-year delivery growth after two consecutive years of declines. Energy storage deployments hit 13.5 GWh against an estimate of 13.3. European markets grew 108% year over year.

The car business is recovering. The market does not care about the car business.

That is the entire story.

The Two-Company Problem

Apply a traditional auto sector multiple of 15x to Tesla's consensus 2026 EPS of roughly $2.00. You get approximately $30 per share. Tesla is trading around $365 to $370 post-selloff.

That means $340 of Tesla's current share price, or roughly 92 to 93% of the entire market cap, is being valued on three bets: Cybercab robotaxis, Optimus humanoid robots, and FSD licensing. None of these generate material revenue today. All three are running behind their original timelines. And all three are being funded by $25 billion in 2026 capex that is three times what Tesla spent in 2025 and will make free cash flow negative for every remaining quarter this year.

Musk is essentially telling investors: the car business is a vehicle for delivering the software. But the software is not yet delivering revenue. So investors are being asked to hold a negative-FCF capex cycle priced on a future business that has not yet proven it works at scale, in a rising-rate environment where Warsh's Fed is pricing 53% odds of a hike before year-end.

That is not a value trap. That is a specific, identifiable risk. The 7.5% drop is the market pricing it correctly, not panicking.

What The Capex Actually Buys

The $25 billion covers six major categories: Cybercab production ramp, Tesla Semi manufacturing, a new Houston Megafactory for Megapack 3 energy storage, Optimus robot production at Fremont replacing Model S and X lines, a lithium refinery and LFP battery factory, and more than doubling AI compute capacity including a dedicated chip fab in Austin.

CFO Vaibhav Taneja was unusually direct on the Q1 earnings call: free cash flow will be negative for the remaining three quarters of 2026. Tesla has $44.7 billion in cash and can fund multiple years of negative FCF without raising equity. So this is not a solvency risk. It is a patience test.

The Cybercab is in production, robotaxi service now covers Austin, Dallas, and Houston with unsupervised FSD. Robotaxi miles nearly doubled sequentially in Q1. Optimus production at Fremont is targeted for late July or August, with Musk declining to detail the V3 design publicly to prevent competitor copying.

Whether any of this delivers material revenue before 2027 is genuinely uncertain. Musk himself acknowledged robotaxi revenue will not be material this year. Wolfe Research projects $250 billion in robotaxi industry revenue by 2035 and Tesla capturing a meaningful share is the bull case. That is a 2035 thesis being priced today in a hawkish rate environment.

The SpaceX Angle Is Real but Overstated

SpaceX buying $269 million worth of Tesla Megapacks for xAI data centers is a genuine commercial relationship, not financial engineering. That is real revenue flowing from Musk's ecosystem into Tesla's energy division. SpaceX's IPO prospectus disclosed that deal, and it validates the Megapack business at scale.

But the broader narrative that the SPCX IPO is a direct TSLA catalyst is weak. The two stocks serve fundamentally different investor bases. SPCX trades at 94x sales on a space and AI infrastructure thesis. TSLA trades at roughly 180x forward earnings on an automotive plus AI plus robotics thesis. These are not competing assets for the same capital in most portfolios.

The more relevant SpaceX-Tesla connection is the Terafab chip fabrication facility in Austin, a $3 billion semiconductor research fab being co-developed with SpaceX handling the initial scale-up phase. If Tesla successfully produces AI chips at Terafab by 2028 or 2029, the compute cost structure for training Optimus and FSD changes materially. That is a 2029 thesis, not a 2026 catalyst.

The July 22 Earnings: What Actually Matters

Deliveries are behind us. The July 22 earnings call is where the market gets the numbers that actually move the narrative.

Three metrics will dominate investor attention.

Automotive gross margin excluding regulatory credits. It recovered to 19.2% in Q1 from 17.9%. It needs to hold or improve in Q2. Any compression will be read as evidence that the volume recovery is coming at the cost of pricing, not in spite of it.

Cybercab production rate. Not the vision. Not the timeline. The actual weekly production number. Musk said the initial ramp follows an S-curve and will be slow. Slow means different things to different investors. Any specific number above 500 units per week would be read as genuine momentum. Below that and the sceptics get ammunition.

Optimus progress. Production at Fremont was targeted for late July to August. Any delay will be significant, because the Fremont factory lines being retooled away from Model S and X production represent a permanent bet on Optimus. If that ramp slips, the market will question the entire manufacturing thesis.

Bull Case vs Bear Case

The Bull Case

The Q2 delivery beat ending two years of consecutive annual declines is a genuine inflection point. Gene Munster called it the first sign the EV winter is ending. European demand up 108% is not noise. The inventory overhang from Q1 is being worked down rather than building. Auto margins are recovering. Energy storage is growing 40% year over year.

The capex spending is being made from a position of genuine financial strength. $44.7 billion in cash. $25 billion in capex still leaves the balance sheet intact. If Cybercab unit economics prove out at scale and Optimus reaches even 100,000 units annually by 2028, the revenue model transforms completely. Wedbush maintains a $550 target. JPMorgan upgraded from Underweight at $145 to Neutral at $475, one of the largest single target revisions ever issued for a large-cap stock. Analyst consensus sits around $403.

The Bear Case

The FCF will be negative for three quarters. Robotaxi revenue is not material this year. The NHTSA has open investigations into Cybercab's design, specifically its lack of backup sensors, which if it leads to a mandatory software requirement could delay autonomous deployment timelines.

93% of the market cap is priced on businesses that are not yet generating cash. In a market where Warsh's Fed is signalling a potential rate hike, multiple compression on speculative high-duration assets is a structural risk, not a cyclical one. JP Morgan's bear scenario sits at $25 per share. Analyst targets span from $25 to $600. When the bull and bear cases are 24x apart, the stock is essentially priced on faith in one man's execution of three simultaneous moonshots at once.

The Tesla Musk premium is real. It is also the primary source of tail risk. Any Musk distraction, whether regulatory, political, or related to his other ventures, has historically hit TSLA harder than fundamental deterioration. That risk does not appear in the valuation models.

The Trade Framework Going Into July 22

At $365 to $370, TSLA is 25% below its December 2025 high of $498. The car business is recovering. The capex cycle is deliberate and funded. July 22 is the next binary.

If Cybercab production rate surprises to the upside and Optimus starts Fremont production on schedule, TSLA has a credible path back toward $420 to $430 before year-end. The delivery beat removes the near-term demand concern. The question is execution on the future businesses.

If gross margin compresses below 19%, Cybercab volumes disappoint, and Optimus slips to Q4, the market will reprice the stock toward $300 to $320 as the consensus view shifts from patient transformation to indefinitely delayed.

The honest position entering July 22: the delivery beat was real and the narrative has genuinely improved. But initiating a full-size position at $370 means you are betting on the AI platform, not the car company. The car company at 15x earnings is worth $30 a share. The rest is faith in a $25 billion capex bet that will not generate material returns until 2027 at the earliest.

Make sure you know which company you are actually buying.

I am not a financial advisor. Trade wisely, Comrades.