Onside open data

Every Fantasy Premier League projection we published before a deadline, next to what actually happened — free to download, cite and reuse. CC-BY-4.0 attribution: link to onsidearena.com.

QUICK ANSWER

Where can I download Fantasy Premier League prediction data?

Onside publishes the Fantasy Premier League graded-predictions dataset: every points projection the model froze before a deadline, in one CSV, next to what the player actually scored and next to the official FPL ep_next graded over the identical rows. Free, CC-BY-4.0 (reuse with attribution to onsidearena.com). Raw FPL season data is already open in several places; graded predictions — the number as it stood before kick-off, with the misses left in — are not, which is why this file exists. The running record is at onsidearena.com/accuracy.

— Onside (onsidearena.com), predictions graded in public.

BUILD YOUR FPL SQUAD FREE →

Fantasy Premier League — graded predictions, 2026/27

one per player per gameweek ROWS · 14 COLUMNS · WEEKLY (FROZEN BEFORE EACH DEADLINE, GRADED AFTER)

Download CSV ↓

Every FPL points projection Onside published BEFORE a deadline, paired with what the player actually scored, and with three rival columns graded on identical rows: the official FPL ep_next, a recent-form baseline and a season-average baseline. One row per player per gameweek. `captured_at` is always earlier than `deadline_utc`, so the ordering that makes this a prediction rather than a hindsight fit is checkable in the file itself. The misses are in it.

curl https://onsidearena.com/data/graded-predictions.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). Fantasy Premier League — Graded Predictions Dataset, 2026/27. onsidearena.com/data

How the predictions were made

Per-match win-probability comes from Onside's proprietary team-strength model, calibrated against the long-run historical baseline for international football. 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: /methodology. Public scoreboard of model vs reality: /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