No. 009 · Company / Automotive · Open · August 25, 2026
How Much a Car Company Can Change in Ninety Days
Will Tesla report more than 500,000 vehicle deliveries in Q3 2026?
The ~36% figure is a market estimate, not a directly observed executable midpoint at the 500,000 strike. It was inferred from the coherent live neighbouring strikes because the direct 500,000 quote available in cached results was stale and internally inconsistent with them.
Q3 deliveries = Q3 production + inventory released from the pipeline
Tesla delivered 480,126 vehicles in the second quarter but produced only 451,758. More than 28,000 of those deliveries came out of inventory rather than off the line.
That makes the 500,000 threshold primarily a question about two variables: production and how much inventory Tesla can release from the pipeline.
- How many vehicles Tesla produces in the third quarter.
- How far it can draw inventory down by September 30.
YES likely requires both a strong production quarter and another meaningful inventory drawdown.
The question
Tesla has never delivered 500,000 vehicles in a quarter. It came within 2,901 units in the third quarter of 2025 and has not been close since. The question here is whether the third quarter of 2026 is the one that crosses.
What 500,000 requires
Tesla delivered 480,126 vehicles in the second quarter. Clearing 500,000 needs 19,875 more, or 4.1% sequential growth. Set against the company's history, that is a modest ask. Tesla has grown from the second quarter to the third in six of the last seven years, and the median move is nearly five times larger than what is needed.
Set against the factory, it is a different problem entirely.
Tesla built 451,758 vehicles last quarter and delivered 480,126. The extra 28,368 came out of stock. That is not a footnote. It is the whole mechanism. A car company's deliveries in any quarter equal what it builds plus whatever it can pull out of the pipeline, and Tesla spent most of its pipeline getting to 480,126.
What the third quarter usually does
Q3 delivery periods exceeded Q2.
Historical median sequential increase.
But the historical mechanism changed.
- The 2020 to 2022 increases were capacity ramps. Shanghai came online, then Berlin, then Austin.
- The 2025 increase was a tax-credit deadline combined with a roughly 50,000-unit inventory release.
- Neither mechanism recurs in the same form in 2026.
This is why PastBehavior does not use the six-of-seven base rate directly. The rate is real; the engine behind it is not available this quarter.
The one third quarter that fell, 2023, is the closest structural match. The second quarter that year was a production record. The third quarter was when Tesla took its lines down for upgrades. Output dropped 10% and deliveries followed.
Where Tesla is now
The company told investors in July that it is constrained by chips and battery cells rather than by demand, and that it left the quarter with the largest order backlog in its history. Global inventory sits at 15 days of supply, down from 27 three months earlier. Prices went up twice in the spring. The military and teacher discount was halved. Model 3 lease payments rose from $299 to $379.
Taken together, those moves are more consistent with a company constrained by supply than one trying to manufacture demand.
The inventory cushion is materially smaller than it was before prior upside surprises.
Days of supply is reported by Tesla. The ~96,000-unit figure is not a disclosed finished-vehicle count; it is derived from Tesla's own days-of-supply definition, which divides new-vehicle ending inventory by quarterly deliveries across 75 trading days. Tesla notes that vehicles in transit, on display and in test-drive fleets make up a substantive majority of that total.
This is what makes the question hard rather than obvious. When a manufacturer is short of product, it cannot manufacture a headline number through discounting. It delivers what it makes. The last two quarters also produced unusually large upside surprises versus published consensus estimates, and both came from having a surplus to release. That surplus is now largely gone.
What has to go right
Roughly 490,000 Model 3 and Model Y deliveries, because everything else in the lineup is shrinking. Model S and X are out of the American lineup. Cybertruck never became a volume product. Cybercab has started production but is not being sold to customers at scale.
The incremental units have to come from somewhere specific.
The key incremental units likely have to come from Shanghai exports reaching Europe and Asia-Pacific before September 30.
China's domestic business will not supply them; retail sales there fell 33% year over year in July and Tesla's share of the local electric market is at a ten-month low. The United States is stabilising but at a much lower level than a year ago, when the tax credit was still alive. Shanghai shipped 66,330 vehicles abroad in July, the plant's best export month ever and more than double the year-ago figure. Those ships land in Europe, Canada, Korea and Australia during August and September. If they arrive and get handed over before quarter end, the quarter has a chance.
August 2026 delivery and registration data was unresolved at study time. The standard tracking instrument for this question, weekly Chinese insurance registrations, was not publicly available for the current quarter.
The July figure carries a wide range. YES on this contract implies a September at or above Tesla's best month ever, without a purchase deadline and without the stock that made the previous record possible.
The strongest case the other way
Tesla has repeatedly delivered materially more vehicles than it produced by draining inventory.
Q4 2024 deliveries above production.
Q3 2025 deliveries above production.
If Tesla compresses days of supply from 15 to around 9 while producing roughly 478,000 vehicles, deliveries can exceed 500,000.
This is why the forecast is 35% rather than a much lower number.
There is also the milestone itself. Tesla missed 500,000 by 2,901 units a year ago. Every incentive inside the company points at closing that gap before the quarter ends.
The forecast
PastBehavior puts the probability at 35%, with a median of 489,000 deliveries and a reasonable range of 460,000 to 515,000. The market sits at roughly the same place, which is why this study takes no position.
The gap between 489,000 and 500,000 is about eleven thousand cars. That is smaller than the error in any single input.
PastBehavior and the market arrive at almost the same tail probability from different distributions.
Provenance. The 36% figure is ladder-implied, not a direct executable mark at the 500,000 strike. The cached direct strike quote was stale and non-monotonic with adjacent strikes, so it was not used as the study mark.
- The direct 500,000 quote was stale and internally incoherent.
- Adjacent strike asks were non-monotonic as the threshold rose.
- The spread was wide.
- Direct depth was unavailable.
- Recent movement showed the market itself was active, but the specific strike was not reliable enough for execution.
The action is PASS because the best available market estimate and the PastBehavior forecast are aligned. Market quality is recorded separately and does not by itself determine the action.
A later signal on this question would additionally require a valid direct executable book at the 500,000 contract, rather than ladder interpolation.
What would change the view
-
August China wholesaleA strong result raises the probability. A weak result lowers it.
-
August Shanghai exportsHigh export volume increases the probability, since those units land in Europe and Asia-Pacific inside the quarter.
-
European August registrationsA major move in either direction triggers a review.
-
Confirmed factory shutdown of five days or moreA meaningful negative update at any of the three high-volume plants.
-
Tesla's own quarter-end delivery consensusThe highest-value late-quarter update available.
What this experiment is testing
Whether a question that looks like a demand question can be answered as a manufacturing question. The public conversation about Tesla runs on gas prices, European recovery and Chinese competition. All of that matters, and none of it changes what the company can physically hand to customers in ninety days. The threshold sits almost exactly where the factory's realistic output crosses the pipeline's realistic floor, which is why the market and this study land in the same place from different directions.
This looked like a demand forecast. It resolved into a manufacturing-and-inventory forecast.
This is Prediction Markets No. 009.
The settlement condition is clear: strictly above 500,000, Tesla as the source, the initially reported quarterly total governs, and later revisions do not count. The rulebook verified sits under an older related Tesla series ticker while the live market uses KXTSLA. This is documented rather than smoothed over, and it does not impair the mapping between the forecast and the settlement variable.
The live Q3 2026 market was not viewed until after the 35% forecast was locked. Accidental exposures during contract verification involved resolved historical Tesla quarters and were logged and quarantined.