| Date | Market | Commodity | Class | Weight | Avg $/cwt | Avg Wt | Head |
|---|
This dashboard pulls structured price data directly from the USDA Agricultural Marketing Service (AMS) MARS API. Understanding how that data is collected, structured, and aggregated will help you interpret the numbers accurately.
All market data originates from the USDA AMS Market News Division and is accessed via the MARS API (marsapi.ams.usda.gov). Data is fetched automatically each weekday evening and stored in Supabase for fast dashboard access.
Texas auction markets tracked:
These markets were selected because they represent significant regional trading hubs and consistently produce enough volume to serve as reliable price benchmarks. USDA reporting operates under Voluntary Price Reporting authority — auction markets choose whether to participate, and USDA must assign a trained federal market reporter to physically attend each sale.
The MARS API is designed for price discovery — helping buyers and sellers understand what cattle of a given type and weight sold for. It organizes data by:
Key limitation: The API focuses on M&L1 frame cattle. Lower frame grades are present in the auction PDF but are not always included as separate API rows. This means API head counts will generally be lower than the PDF's "Total Receipts" figure.
1. Frame grade filtering — The API primarily returns M&L1 rows. M&L 1-2 cattle run through the same ring but are not consistently represented in the API's machine-readable rows.
2. Unsold & passed cattle — The PDF "Total Receipts" includes cattle that were passed (did not reach the minimum bid) or withdrawn before sale. Since these have no transaction price, the API has nothing to record.
3. Commodity filter — The Market Comparison tab filters to one commodity at a time, so head counts reflect only the selected category, not all cattle at the auction.
Example — San Angelo, April 2026: The PDF reports "Total Receipts: 373" (307 Feeder + 62 Slaughter + 4 Replacement). The MARS API returns ~305 head across all commodities. All three figures are technically correct — they simply count different things.
All price figures are weighted average prices — each lot's price is weighted by its head count before averaging. A lot of 30 steers carries more weight in the average than a lot of 3 steers. This is the same methodology USDA uses in the PDF reports.
Prices are expressed as $/cwt (dollars per hundredweight) for Feeder and Slaughter Cattle, and as $/head for Replacement Cattle. Weekly averages group all auction dates within the same Monday-through-Sunday week.
Despite the limitations above, this data is highly reliable for its primary purpose — tracking price trends and market comparisons over time. Because the same subset of cattle is captured consistently week after week, the trend lines accurately reflect how prices are moving. Use this for: