October retail sales, before the print.
Advance retail and food services sales and the control group that feeds GDP. The call uses light-vehicle unit sales (out early each month), pump prices for gas-station receipts, and momentum.
What September 2026 told us
The last release against what was expected, what drove it, and what it changes.
The print
| First print | Now | Consensus | Model | Key Data call | Surprise | |
|---|---|---|---|---|---|---|
| Retail sales | +0.8% | +0.77% | +0.3% | +0.21% | +0.30% | +0.5 pt |
| Control group | +0.14% | +0.14% | – | +0.03% | – | – |
What it showed
- September 2026 retail sales rose +0.77% against a consensus of +0.3%; the control group that feeds GDP rose +0.14%.
- Largest contributions: gasoline stations +55.1 bp, food services & drinking +13.9 bp, nonstore retailers +4.8 bp.
- Sales are 6.3% higher than a year earlier in nominal terms.
- The Chicago Fed's card-spending tracker (CARTS) had sales ex autos at +0.43% for September 2026; Census printed +0.90%.
What drove it
Contribution to the headline change, basis points.
Analyst notes
Strong headline, soft underneath. Retail sales rose 0.8%, well above the 0.3% consensus, but more than two-thirds of the gain came from gasoline stations, where sales jumped 6.4% on higher pump prices. That is prices, not volumes.
The control group tells the real story. Sales excluding autos, gasoline, building materials and restaurants rose just 0.1%, a soft input for third-quarter consumer spending in GDP. Restaurants (+1.1%) were a bright spot.
October. With pump prices flat to lower and vehicle sales not yet reported, the model expects a modest headline gain near 0.2%, with the control group around 0.3–0.4%.
Where October 2026 should land
Odds of each outcome, sized by the model's own recent misses with fat tails for surprise months.
Retail sales model +0.25%
Control group model +0.37%
Vehicle sales for October 2026 are not out yet, so this call assumes they hold at last month's pace.
Track record
Every month the model is refit using only data available at the time, then scored against the number first published, the same yardstick used for the consensus.
| Period | Releases | Model miss | Key Data call miss | Consensus miss | Exact rounded print: model / call / consensus | Model closer / further | |
|---|---|---|---|---|---|---|---|
| Since 2023 | 45 | 0.359 | 0.318 | 0.318 | 7% / 4% / 4% | 21 / 24 | Even |
| Since 2019 | 77 | 0.454 | 0.378 | 0.378 | 8% / 6% / 6% | 32 / 45 | Even |
| Since 2014 | 130 | 0.415 | 0.332 | 0.332 | 7% / 8% / 8% | 51 / 79 | Even |
How it works
Headline: the twelve-month pace plus the month's change in pump prices and in BEA's light-vehicle sales rate, the two biggest swing factors in nominal receipts, and new and used vehicle prices from the CPI once it is out (a few days before retail sales). It learns from advance prints, the number markets trade on. Until vehicle sales for the month are out (around the third business day), the call assumes they hold steady.
Control group (excluding autos, gasoline, building materials and restaurants): momentum only. Card-spending data that professional forecasters use is not public, so the consensus has an edge here and the Key Data call defers to it.
The Key Data call: once Wall Street's consensus is published, the model's view is combined with it. The combination starts from the consensus and moves toward the model by an amount learned from past releases, also weighing last month's surprise and revision. It is used only while it has beaten the consensus over the previous three years of releases, scored out of sample; otherwise the call is the consensus.