Football Charts API Review 2026

Football Charts is a soccer stats site that puts a model's match probabilities next to the betting market's prices, and it opens the data behind its charts as a REST API and an MCP server. One person builds and runs it, and the site publishes every settled prediction, losses included.

The API covers 98 leagues in 44 countries. That list goes deep: England's National League, Germany's 3. Liga, Spain's RFEF groups, and five women's leagues, including England's WSL and Germany's women's Bundesliga. Data updates once a day, around 06:00 UTC, so it suits models, dashboards and research. It does not carry live scores or bookmaker odds.

Each endpoint takes a league code, so a Premier League call (premier) and a Bundesliga call (germany1) differ by one word. What comes back:

• Results: full-time and half-time scores, the first-goal minute and a UTC kickoff time.
• Tables: standings with form, clean sheets, over 2.5 rates and expected points, plus a "luck" view that shows who sits above or below their numbers.
• Fixtures: upcoming matches with the model's probabilities.
• Season projections: title, top-four and relegation chances from 10,000 simulated seasons, run daily.
• Goal timing: goals per 15-minute window, by league or by team.

Paid keys add a stats object to each finished match: shots, passes, possession, corners, cards and saves, plus xG where the source has it. The docs put shots at about 60 leagues and xG at about 25. There is no player data. For live prices, see our soccer odds API guide. For open history in one file, Football Charts also publishes a free dataset, which sits on our free sports datasets list.

The probabilities come from a Dixon-Coles scoreline model, a standard method that predicts goals for each side and builds every market from one score grid. Since October 28, 2026, a machine-learning layer adjusts it with recent form and shots. Because 1X2, totals and both-teams-to-score all come from the same grid, they always agree with each other.

The site is honest about the model: it says the model does not beat the bookmaker market, and the /track-record/ endpoint returns the settled public ledger so you can check. Treat the numbers as a clean baseline for your own model, not as picks.

The free tier gives 25 requests a day with no key, or 200 a day with a free key once you confirm your email. It covers all 98 leagues for the current and previous season, for personal use, with attribution.

• Personal, €9 a month: 5,000 requests a day, every season back to 2020 and the per-match stats. Also sold as a €49 pass that runs through July 2027.
• Commercial, €39 a month: 25,000 requests a day, and the license you need for anything other people use, such as an app, site or bot, paid or free.

The odds history is a separate €199 download in CSV and Parquet, not part of the API: opening and closing prices from about three bookmakers a match, 2020-21 to 2025-26. Check the terms before you ship: no plan allows you to resell or republish the raw data.

You can start without one: the API answers 25 requests a day per IP address with no key. For more, enter your email on the developers page and open the confirmation link. The key works at 100 requests a day until you confirm, then 200. Send it as Authorization: Bearer fc_your_key. A paid plan upgrades the same key, so your code does not change.

This sample reads the Premier League table with no key and prints the top five with points and expected points. We ran it on November 12, 2026.

import requests

BASE = "https://footballcharts-backend.onrender.com/api/v1"

resp = requests.get(f"{BASE}/leagues/premier/table/", timeout=30)
resp.raise_for_status()
data = resp.json()

print(data["season"], "updated", data["table"][0]["updated_at"])
for row in data["table"][:5]:
    print(row["position"], row["team"], row["points"], "pts,", row["expected_points"], "xPts")

Swap table for results, fixtures, projection or goal-timing, and premier for any code that GET /leagues/ returns. With a key, add headers={"Authorization": "Bearer fc_your_key"}, and add ?format=csv for a CSV answer.

Limits count per day: 25 per IP address with no key, 200 per email address on a free key (60 a minute), 5,000 on Personal and 25,000 on Commercial. Every answer carries x-ratelimit-limit and x-ratelimit-remaining headers. The developers page lists every endpoint, and the API describes itself as an OpenAPI 3 document at /api/v1/openapi.json, which Postman and code generators read.

There is no client SDK. The MCP server is the other way in: add https://mcp.football-charts.com/mcp as a custom connector in Claude and you get ten read-only tools, 50 calls a day without a key. The open-source server also installs with npx -y footballcharts-mcp. See our sports MCP servers guide for others.

• Modelers and bettors who want clean results, goal minutes and a public baseline model to test against, across leagues most APIs skip.
• Analysts and spreadsheet users who want tables and results in CSV or straight into Google Sheets.
• AI assistant users who want to ask Claude about any of 98 leagues with no setup.

If you need live scores, lineups, player stats or odds over an API, compare the sports data APIs or the full directory.

A cheap, honest source for soccer results and probabilities with unusually deep league coverage. The free tier is enough to build and test a model on two seasons, and €9 opens history back to 2020. Daily updates and no player data rule it out for live apps, but for lower-league research and model work it is hard to beat at the price.

Last updated: 12 November 2026 · How providers get listed