Model-Year Changeover Tracker
How fast the incoming model year of one model is replacing the outgoing one, month by month, and where the leftovers are still selling by state.
This is the technical reference. For the price, a rendered sample and answers to common questions, see the Model-Year Changeover Tracker product page.
At a glance
| Price | $4.99 per run. |
| Tool name | generate_model_year_changeover |
| Scope | reports:model-year-changeover-tracker |
| Built for | Automaker, Small Dealer |
What it does
The model-year changeover tracker for one model, from Sold Summary API v2 (the model-year dimension, so one 12-month call is the whole sell-down curve): units, share of the model's units and the gap to sticker per model year per month over the window (12 months by default), with at most three model years drawn and the rest pooled as Older; the incoming model year (the newest with at least 1 percent of the latest month's units), the outgoing one before it, the first month the incoming year appeared and the crossover month it first outsold the outgoing year; and by state for the latest month, where the outgoing year's leftovers are still selling, with the outgoing year's share of the state's sales, its gap to sticker, days to sell and median asking price. Three findings are written from the numbers: how far the takeover has gone and when it crossed, how the two years price against sticker, and where the leftovers are. New vehicles only. From the Sold Summary API: a sale is inferred when a listing leaves a US dealer website, every price is the advertised asking price at exit (never a negotiated transaction price), and thin rows of old model years recorded as new are pooled as listing-data noise. Use when a launch, allocation or sell-down team asks how fast the new model year is taking over, when it crossed 50 percent, how the outgoing year is priced against the incoming one, or which states still carry leftovers. Do NOT use for which model years are thinning on the used side (use Late-Model Cohort Thinning), for one model's trims (use the Trim Contenting Report), or for a state discount map of the whole brand (use the Regional Discount & Scarcity Map). PRICING: $4.99 per run.
Arguments
| Argument | What it is | Example | |
|---|---|---|---|
make |
Required | Make of the model (e.g. Subaru). | Subaru |
model |
Required | The model (e.g. Outback), as the sold data spells it. | Outback |
months |
Optional | Months in the curve, 2 to 12, ending at the latest complete data month (12 by default). A full changeover takes four to six months, so 12 shows the whole curve. | 12 |
What comes back
- The changeover curve: units, share of the model's units and the gap to sticker per model year per month, at most three model years drawn and the rest pooled as Older
- The incoming and outgoing model years, the first month the incoming year appeared, the crossover month it first outsold the outgoing year, and the phase of the changeover
- Leftovers by state for the latest month: the outgoing year's units, its share of the state's sales, gap to sticker, days to sell, median asking price and the priced sample; the fifteen largest states in the ladder, every state in the table and the CSV
- Three findings from the numbers: how far the takeover has gone and when it crossed, how the two years price against sticker, where the leftovers are
- Thin rows of old model years recorded as new pooled as listing-data noise and said so
- Next steps: sell-down support for the leftover states, the Trim Contenting Report, a monthly monitor in changeover season
Asking for it in a chat client
Model-year changeover tracker for the Subaru Outback
How fast is the 2026 RAV4 replacing the 2025 by state?
Running it from a backend
Grant an agent the reports:model-year-changeover-tracker scope, then call the tool over MCP:
{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "generate_model_year_changeover",
"arguments": {
"make": "Subaru",
"model": "Outback",
"months": 12
}
}
}
See Server-side access for the token exchange and how to read the response.