Goldman Sachs cut its estimate for Tesla's third-quarter deliveries to 435,000 vehicles on 17 September 2026, down from 490,000. The fourth-quarter number came down in the same note, from 515,000 to 475,000.
What actually changed, and what did not
Analyst Mark Delaney's reasoning is that deliveries are tracking below consensus across Tesla's three largest markets at once — the United States, China and Europe. Export strength, he argues, is not enough to offset that.
What did not change is the recommendation. Goldman stays Neutral on the stock with a $360 price target. Only the volume assumptions moved, which is the part that matters to anyone reading this as a demand signal rather than a trading note.
| Quarter | Previous | Revised | Change |
|---|---|---|---|
| Q3 2026 | 490,000 | 435,000 | −55,000 |
| Q4 2026 | 515,000 | 475,000 | −40,000 |
That is 95,000 vehicles taken out of the second half of the year.
Where the number sits against everyone else
A single bank's estimate is only useful next to the alternatives, and on this one the spread is unusually wide.
| Source | Q3 2026 deliveries |
|---|---|
| Goldman Sachs | 435,000 |
| Visible Alpha consensus | 456,000 |
| Kalshi prediction market | around 480,000 |
Goldman is roughly 21,000 below the analyst consensus and around 45,000 below where prediction-market traders have settled. When a sell-side desk and a prediction market disagree by a tenth of the quarter's volume, the honest reading is that nobody outside Tesla knows yet.
The European part of the claim
Europe is one of the three markets Delaney names, and it is the one European readers can partly check for themselves. August registration data across the continent was genuinely mixed rather than uniformly weak — some markets up sharply year on year, others down by comparable margins — which is what happens when incentive changes and shipment timing land in different months in different countries. Tesla's China performance was covered separately in Tesla's August retail and export figures.
The quarter also has a structural quirk that forecasts routinely trip over: Tesla ships to Europe early in a quarter and delivers late, so September registrations in individual European countries are a poor guide to the global number until the quarter closes.
What to do with this
Very little, immediately. A delivery forecast is a bank's opinion about a number Tesla will publish itself within days of the quarter ending on 30 September. If you are waiting on a car, nothing here changes your delivery date. If you are deciding when to order, the only actionable part is indirect: quarter-end is historically when inventory cars and incentives appear, and a quarter tracking below expectations tends to produce more of them, not fewer.
The forecast is worth remembering mainly so the actual result can be judged against something. In early October there will be a real figure, and 435,000 is one of the marks it will be measured against.