Use cases & endpoints
WASDE report data API — supply and demand estimates
WASDE is a scheduled monthly shock. The report sets out world agricultural supply and demand estimates, and grain markets respond to the revisions in it — the change from last month's estimate, not the absolute level, is what trades.
As a data source it is published as a document with tables, revised repeatedly over a marketing year, and only meaningful with its release attached. A database that keeps the latest figures answers what the USDA now believes and cannot answer what the market learned on release day, which is the question anyone studying the report is actually asking.
Hindcast stores each release as its own vintage with line items structured, so the diff between consecutive reports is a query rather than a spreadsheet exercise. Release timestamps travel with the values, making event studies against price reactions straightforward and honest.
hindcast — session
previewhindcast> what did the latest report revise versus the previous one?
→ tool: wasde_release { commodity:"...", release:"latest", diff:"previous" }
✓ changed line items with both vintages side by side
source: USDA WASDE, each monthly release retained
What's covered
- Every monthly release retained as its own vintage
- Line items structured rather than left as document tables
- Release timestamps attached for event-study alignment
- Vintage-to-vintage diffs computed directly
- Marketing-year context preserved on each estimate
- Linked to positioning and price data through shared identifiers
Join the waitlist — this page's dataset is already selected.
Questions
Why not just use the final revised estimates?
Because nobody traded the final estimate. The revision path is the information: prices responded to each release as it landed, and a study using end-state numbers is analyzing a report that never existed in that form.
Can I line the report up with price reactions?
That is the intended use. With release timestamps on the report and end-of-day prices on the same identifiers, a before-and-after comparison needs no manual alignment.
Related
Join the waitlist. Vote the roadmap. First in, first served.