Onside.
OPEN DATA · CC-BY-4.0
3 datasetsCSV + JSONCC-BY-4.0

Onside open data

Every World Cup 2026 prediction, fixture and champion-probability sample we publish — free to download, cite and reuse. CSV for spreadsheets, JSON for code. CC-BY-4.0 attribution: link to onsidearena.com.

World Cup 2026 — model predictions × results

72 ROWS · 15 COLUMNS · LIVE (EVERY MATCH UPDATE)
Download CSV ↓

Every group-stage fixture with Onside's pre-match win-probability split (home / draw / away), plus actual scoreline and verdict once each match is played. Refreshes live during the tournament.

JSON API → /api/v1/wc/predictcurl https://onsidearena.com/data/predictions.csv

World Cup 2026 — champion probability board

48 ROWS · 10 COLUMNS · HOURLY
Download CSV ↓

Top-32 champion probabilities from the 5,000-run Monte Carlo simulator, with reach-round probabilities (R16 / QF / SF / Final). Auto-refreshes hourly and locks in real results as priors.

JSON API → /api/v1/wc/championscurl https://onsidearena.com/data/champions.csv

World Cup 2026 — 104-match schedule

104 ROWS · 9 COLUMNS · STATIC (24H CACHE)
Download CSV ↓

Full FIFA World Cup 2026 schedule: all 72 group + 32 knockout fixtures with venues, kickoff times (UTC), and matchday numbers. Cross-checked against Sky Sports + Wikipedia + FIFA portal.

JSON API → /api/v1/wc/fixturescurl https://onsidearena.com/data/fixtures.csv

License + attribution

All datasets are licensed CC-BY-4.0. You can use them for any purpose — commercial or non-commercial, research, journalism, fantasy communities. We ask one thing: include a link back to onsidearena.com or the specific dataset URL when you publish.

Suggested citation: Onside (2026). World Cup 2026 — Model Predictions Dataset. onsidearena.com/data

How the predictions were made

Per-match win-probability comes from a logistic opponent-rating model with four signals (FIFA rank, Premier League squad footprint, host advantage, confederation strength). The champion board is a 5,000-run Monte Carlo over the full tournament bracket, using sampled goals from a Poisson process calibrated to the per-match probability.

Full methodology, formulas and calibration receipts: /world-cup-2026/methodology. Public scoreboard of model vs reality: /world-cup-2026/model-record.

Coming next

  • · Kaggle mirror with notebook starter pack (June 11+)
  • · Hugging Face Datasets mirror for ML workflows
  • · Per-player FIFA Fantasy projections CSV (post-MD1)
  • · Post-tournament backtest dataset with calibration analysis

Building something with this?

We'd love to see what you make. [email protected]. If you want your AI assistant to query this directly, install our MCP server:

npm install -g onside-football-mcp