Carvana heads into its third-quarter report with a story that has two very different halves. Demand looks abundant, unit growth is still running near 40%, and management has finally put a number on its 2026 ambitions with adjusted EBITDA guidance of $2.7 billion to $3.0 billion. Yet profit per car has now shrunk year over year for three straight quarters, and the stock has drifted lower even as the company keeps setting volume records. The report on October 28 should show which half of that story deserves more weight.

The Street expects earnings of $0.48 per share on revenue of $7.62 billion. That revenue figure would represent roughly 35% growth from a year ago and a modest step up from the $7.38 billion reported last quarter. Earnings tell a different tale: the consensus implies a decline of more than 50% from the $1.03 posted in the year-ago period, though it would edge above last quarter's $0.42. Revenue growth also deserves some care in interpretation. Management flagged that a gross-revenue accounting benefit rolled off after the second quarter, so reported sales growth should now track closer to underlying unit growth. A mid-30s revenue increase therefore implies retail units growing at roughly that pace, a touch slower than the 38% delivered last quarter, which itself was down from 40% the quarter before.

That makes inventory the operational swing factor. Last quarter, management showed regional data in which the Midwest and Northeast grew inventory 57% and sales 54%, while the West and Southeast added just 17% more inventory yet still lifted sales 30%. The message was that demand is not the constraint; supply is. But the company also admitted inventory undergrew sales during the second quarter and said it would lean on heavier advertising in the third quarter to compensate. Evidence that inventory has caught up, alongside unit growth holding in the high 30s, would validate the flywheel argument. A sharper deceleration paired with elevated marketing spend would suggest the company paid more to sell fewer incremental cars than hoped.

Margins are the other half of the test. Adjusted EBITDA margin slid to 10.4% last quarter from 12.4% a year earlier, and the year-ago third quarter sets a demanding 11.3% comparison. Pressure came from nearly every line: lower retail profit per unit from tariff lapping and fuel, softer wholesale results, and a steep drop in other gross profit as Carvana passed about 100 basis points of rate savings back to customers. Management framed much of this as deliberate reinvestment. Investors should look for signs that per-unit compression is at least stabilizing, and that the recovered reconditioning efficiency is showing up in costs. The full-year guide implies second-half EBITDA roughly flat with the first half, which analysts pushed back on as conservative. A quarter that tracks toward the upper end of the range would make that skepticism look justified; anything implying the low end would reinforce worries about margin drift.

The mood has cooled noticeably. Sentiment has swung from modestly bullish ahead of the last report to modestly bearish now, and the stock is essentially flat since then while the S&P 500 gained about 4%. At $58.71, shares trade well below the 200-day moving average near $70.53, sit close to the post-earnings low of $56.12, and have slipped beneath the prior quarter's trough. Expectations, in other words, look more subdued than the operating trajectory alone might suggest.

The central question is whether Carvana's margin giveback is buying durable growth. If inventory has been rebuilt, unit growth is holding up, and EBITDA per unit shows signs of bottoming, the confident scaling narrative remains intact. If volume slows while margins keep eroding, the market's recent skepticism will look well placed.