Build against a production football intelligence contract, not a loose demo surface. The playground is for request validation, the document is for schema discipline, and the API outputs are designed to move directly into scouting tools, report builders, and internal analytics systems.
https://footballperformanceapi.site Use one production host across exploration, schema reference, and implementation so endpoint behavior does not drift between testing and release.
The fastest integration path is to prove the request in the playground, confirm the response contract in Redoc, then copy the same endpoint and payload assumptions into your service layer.
Exercise real endpoints with the same request contract your application will use. Use it to confirm authentication, payload shape, response fields, and edge cases before writing integration code.
Reference DocsUse the generated schema as the source of truth for parameters, response objects, validation rules, and the data types returned by each football intelligence module.
Treat the playground as a contract rehearsal. Check required fields, sample realistic player inputs, inspect error responses, and lock the final payload before connecting the API to a dashboard, report generator, or batch workflow.
The API is organized around outputs that remain useful after the first request: player scoring, role interpretation, custom prediction, and report-ready structures for downstream products.
Consume normalized player performance scores that are stable enough for dashboards, scouting shortlists, automated reports, and downstream ranking logic.
Retrieve role-cluster results and similarity signals so player comparison is based on behavior patterns instead of only nominal position labels.
Submit private or long-tail player feature sets to generate an initial rating and role interpretation when the public database is incomplete.
Connect structured results to player reports, model validation pages, comparison tools, and internal scouting views without reformatting model output by hand.
Customized prediction is for cases where the player matters before the database is complete: academy players, smaller leagues, private scouting files, trials, or internal training data.
Start with a representative payload, check how missing or low-density features behave, then reuse the same contract in your scoring workflow. This keeps custom player evaluation auditable instead of becoming a spreadsheet-only process.
Open the playground, test a realistic request, then use the API document to lock down endpoint behavior before production work.