After Wednesday's close (7/22), Tesla reported Q2 2026. Headline "beat," a print that fell apart the moment you read past the top line. As of midday Thursday the stock is down ~13% to $324.64, erasing roughly $200 billion of market cap in a single session and marking its worst post-earnings reaction in over a decade.
| Metric | Q2 2026 | Estimate | Read |
|---|---|---|---|
| Revenue | $28.24B (+25.5% YoY) | ~$25.7–26.7B | Beat (+5.7% to +10.5%) |
| Deliveries | 480,126 (record, +25% YoY) | Below actual | Genuine beat |
| Adjusted EPS | $0.33 | ~$0.49–0.54 | Miss ~33–39% |
| Gross margin | 16.8% | 19.4% | Big miss |
| Auto GM ex-credits | 16.3% (from 19.2% Q1) | Below bear case | Structural erosion |
| Operating margin | 1.4% (from 4.1% YoY) | — | Op income −57% to $398M |
| Free cash flow | −$1.09B | — | Negative vs +$1.44B in Q1 |
| Capex | $5.79B (+142% YoY) | $6.59B | FY26 guided >$25B, rising |
| Regulatory credits | $146M | — | From $380M (Q1) / $439M (YoY) |
Per Tesla CFO Vaibhav Taneja on the call, GAAP net income was "positively impacted by a mark-to-market gain of $1 billion on our SpaceX holdings," partly offset by ~$300M in FX losses and ~$100M on Bitcoin. Investing.com's Thomas Monteiro did the math bluntly: "Roughly $0.22 of the $0.32 GAAP EPS came from an unrealized $1 billion gain on Tesla's recently acquired SpaceX stake — a paper markup rather than earned profit." Strip it out and the operating business earned roughly a dime a share on $28B of revenue. This is the same trick Alphabet used the same night (an equity-gain-flattered print) — but Alphabet has an 80%+ growth cloud business and real free cash flow underneath it. Tesla does not.
For years, near-100%-margin regulatory-credit sales were a quiet profit crutch. This quarter they collapsed to $146M from $380M last quarter and $439M a year ago — because the Trump administration cancelled the federal mechanism that forced other automakers to buy them. This is not a timing blip; it is a permanent hole in the margin structure that Tesla now has to fill with a car business whose gross margin (ex-credits) is falling, not rising.
Tesla is now visibly two companies stapled together with a single, extreme valuation:
The entire $1.4 trillion valuation rests on Tesla #2. And on this call, Musk himself managed those expectations down.
I read the full Q2 call (prepared remarks + Q&A). The market didn't sell the numbers alone — it sold what Musk and Taneja said. Three quotes did the damage:
"We should be spending on CapEx as fast as we can — as fast as we can without it being too wasteful." — Elon Musk
Taneja translated that into hard guidance: FY26 capex >$25B, "will grow for the next two or three years," and Tesla is "arranging debt facilities that will give us the capacity to borrow up to $30 billion" to accelerate it. A company earning a dime of real EPS is levering up to fund a multi-year negative-FCF build. That is a religious bet, not a financial plan.
"Optimus will follow the normal S-curve of a manufacturing ramp, but the initial portion of the S-curve will be quite flat and long because of the newness of the parts... there is no supply chain." — Elon Musk
This is the single most important sentence in the call. The most-hyped catalyst — the one analysts assign the most speculative value to — was explicitly de-risked by the CEO to "flat and long." Piper Sandler admitted it values Optimus at $0/share today. When the bull case is "trust the flat part of the S-curve," near-term buyers evaporate.
"Now, it's quite unclear what SECULAR driver investors are focusing on for the near-term." — Evercore ISI
Evercore named the real problem: Tesla has always traded on the next shiny object (FSD → robotaxi demos → Optimus SoP → Terafab). This quarter, every shiny object came with a "be patient" caveat, and the AI-capex-skepticism regime that hit Alphabet the same night left no appetite for optionality without ROI.
Tesla trades at a trailing P/E of ~360× and forward ~370× (EPS TTM ~$1.09), with an ROE of 4.8% and net margin in the low single digits. On any auto or even any-industrial framework, this is indefensible; it is purely an options premium on Tesla #2.
Analyst reactions post-print (note the enormous dispersion — itself a red flag that nobody can anchor fair value):
| Firm | Rating | Target | Note |
|---|---|---|---|
| Evercore ISI | In Line | $300 | "Unclear what secular driver" |
| Morningstar (FV) | Narrow moat | $400 | Fair-value estimate |
| JPMorgan | Neutral | $445 (from $475) | "Range-bound near-term" |
| Mizuho | Outperform | $450 (from $480) | Physical-AI leader |
| Piper Sandler | Overweight | $500 | Optimus valued at $0 today |
| RBC | Outperform | $500 | Robotaxi on schedule |
| Stifel | Buy | $508 | Margin compression "modestly negative" |
Consensus 12-month target was cut to ~$392 (implying only ~5% upside from Wednesday's close before today's crash). The bear SOTP camp is far lower: William Blair pegged the core auto business at ~$30/share; independent SOTP work this year has landed near $127. The truth sits between the religion and the nihilism.
Oversold does not mean safe: RSI has been oversold for days and price kept falling. But a violent 13% down-day near a well-defined 52-week low is exactly where dead-cat bounces start — which is why I'd rather sell strength than short weakness here.
(a) AI-capex unbundling. The same night, Alphabet fell ~6.5% (to $319.81) on a capex hike to $195–205B despite 80%+ cloud growth. The market's new rule: it rewards AI hardware (SMCI, Micron, AMD) and punishes platform spenders who can't show near-term ROI. Tesla is the worst-positioned name in that frame — unlike Alphabet, its capex has no current cash engine underneath it. JPMorgan's technical desk warned this "could shift from a rotation within the AI theme into a more pronounced position unwind." With MSFT/META (7/29) and AAPL/AMZN (7/30) on deck plus FOMC 7/28–29, the setup is fragile.
(b) The oil irony. Brent topped $100 and WTI cleared $90 today on Hormuz/Red Sea escalation (Goldman: $120 in Q4 if disruptions persist). Counterintuitively this is a modest tailwind for Tesla demand — $4+ U.S. gasoline and ~$9/gal in Germany pushed EV adoption in H1. But it is nowhere near enough to offset a margin structure in decline, and higher rates (which oil-driven inflation reinforces) directly raise Tesla's financing-subvention costs, a headwind Taneja called out explicitly.
Bottom line: This is not a broken company — robotaxi and Optimus are real long-duration options. It is a broken near-term risk/reward: 360× earnings, a 1.4% operating margin, a paper-flattered profit, a permanent credit hole, and a CEO telling you the payoff curve is "flat and long." Patience beats conviction here. I want it $60–$90 lower before the optionality is worth paying for.
TSLA (called 7/22): I flagged Tesla as "CAUTION/AVOID short-term — cash burn trend (FCF −$1.1B), margin compression 16.8%, no formal entry." That call is vindicated: the stock fell 13% the very next session on precisely those factors. I'm keeping the discipline — still no long.
GOOGL (called 7/22, HIGH CONVICTION buy-the-dip $325–338, tgt $400): Honesty check — GOOGL is $319.81 today (−6.5%), just below my entry zone as the capex-skepticism wave hit harder than expected. The thesis (record cloud, real FCF, backlog) is intact and my invalidation (weekly close <$310) has not triggered, but it's underwater and on a short leash. I'd rather be early and wrong-footed by a week than chase — but I acknowledge the entry was early.