Historical World Cup Data API: Every Match from 1930 to 2022

The 2026 FIFA World Cup will be the 23rd edition of the tournament. The first 22 — from 1930 in Uruguay to 2022 in Qatar — contain ~900 matches, 22 champions, 22 Golden Boot winners, and the data underpinning every "who's the GOAT?" debate.

If you're training a prediction model, building a tournament retrospective, writing a data-journalism piece, or feeding historical context into a live ticker app, you need fast access to this archive. This guide covers what's available and how to pull it.

What historical World Cup data is available

TheStatsAPI provides:

• Every match from 1930-2022 with final scores
• Tournament metadata per edition: host, winner, runner-up, 3rd/4th place, top scorer, total goals, total matches, total teams
• Knockout-stage results including extra time and penalty shootouts
• Top scorers per tournament (Golden Boot history)
• Hosting history by country and city
• Format changes documented (1934 single-elim, 1982 expansion to 24, 1998 expansion to 32, 2026 expansion to 48)

The depth varies by era. Modern World Cups (post-1990) have lineups, substitutions, and detailed events. Early editions (1930-1950) are primarily final scores and tournament summaries — that reflects what was actually recorded at the time, not API gaps.

Quick fetch: every edition's summary

curl 'https://api.thestatsapi.com/api/football/competitions/{COMPETITION_ID}/seasons' \
  -H 'Authorization: Bearer YOUR_API_KEY'

Returns one summary object per World Cup edition with the winner, runner-up, top scorer, and totals. Useful for building a historical timeline component.

Fetch all matches from a single tournament

curl 'https://api.thestatsapi.com/api/football/matches?competition_id={COMPETITION_ID}&season=2018' \
  -H 'Authorization: Bearer YOUR_API_KEY'

Returns the 64 matches of the 2018 World Cup in Russia. Swap season for any year between 1930 and 2022.

Train a prediction model on historical data

A typical workflow for a World Cup 2026 prediction model:

import requests
import pandas as pd

# Pull every World Cup match since 1990 (more reliable data)
all_matches = []
for season in range(1990, 2023, 4):  # 1990, 1994, 1998, ..., 2022
    r = requests.get(
        'https://api.thestatsapi.com/api/football/matches',
        params={'competition_id': COMPETITION_ID, 'season': season, 'per_page': 100},
        headers={'Authorization': 'Bearer YOUR_API_KEY'},
    )
    all_matches.extend(r.json()['data'])

df = pd.DataFrame([{
    'season': m['season'],
    'home': m['home']['name'],
    'away': m['away']['name'],
    'home_score': m['home_score'],
    'away_score': m['away_score'],
    'stage': m['stage'],
} for m in all_matches])

# Use this to train an Elo, Poisson, or Dixon-Coles model
print(f"Loaded {len(df)} historical World Cup matches")

That's ~500 matches between 1990 and 2022, enough to fit a basic Poisson model with team-strength priors.

Compute career World Cup stats per player

# All goals by Cristiano Ronaldo across his World Cups
goals = []
for season in [2006, 2010, 2014, 2018, 2022]:
    r = requests.get(
        f'https://api.thestatsapi.com/api/football/players/cristiano-ronaldo/stats',
        params={'competition_id': COMPETITION_ID, 'season': season},
        headers=headers,
    )
    stats = r.json()['data']
    goals.append({'season': season, 'goals': stats.get('goals', 0), 'apps': stats.get('appearances', 0)})

total = sum(g['goals'] for g in goals)
print(f"Ronaldo: {total} goals across {len(goals)} World Cups")

Top scorers (Golden Boot) by year

Fontaine's 13 goals in 1958 remains the single-tournament record, set in six matches. Available via:

curl 'https://api.thestatsapi.com/api/football/players/stats?competition_id={COMPETITION_ID}&season=1958&order_by=goals&limit=10'

Hosting history

Every country that has hosted the World Cup, with the year and final venue:

curl 'https://api.thestatsapi.com/api/football/competitions/{COMPETITION_ID}/hosts'

The 2026 World Cup is the first three-host tournament (USA, Canada, Mexico). Estadio Azteca becomes the first stadium to host matches at three different World Cups (1970, 1986, 2026) — useful trivia for your match-page sidebar.

Use cases for historical World Cup data

• Prediction models — Elo, Poisson, Dixon-Coles, neural-net training
• Editorial / journalism — "all-time top scorers" listicles, head-to-head retrospectives
• Match-page sidebars — "the last time these teams met at a World Cup..."
• Anniversary content — "On this day in 1986, Maradona..."
• Fantasy game scoring backtests — verify your rules against past tournaments
• Wagering models — historical performance feeds into pre-tournament outright odds
• Education — build datasets for sports analytics courses

Caching guidance

Historical data is fully static. Cache aggressively — daily or weekly is fine. Pull once at build time and bake into your static export if you're using Next.js.

Frequently Asked Questions

How far back does the World Cup data go?

To the inaugural 1930 World Cup in Uruguay. Every edition from 1930 to 2022 is covered. Depth varies — modern editions (post-1990) have detailed lineups and events; early editions have final scores and tournament summaries.

What was the highest-scoring World Cup?

The 1958 World Cup in Sweden featured 126 goals in 35 matches (3.6 goals per match). The 2022 World Cup had the most total goals (172) but across more matches.

Who is the all-time top scorer at the World Cup?

Miroslav Klose (Germany) with 16 goals across 4 tournaments (2002, 2006, 2010, 2014). Ronaldo Nazário (Brazil) is second with 15.

Can I get xG for historical World Cup matches?

xG was not tracked at most pre-2014 World Cups in any meaningful way. From 2018 onwards, expected goals are available. For earlier matches, you'll need to estimate from box-score events.

How do I build a "what happened on this day" widget?

Filter the matches endpoint by month/day across all seasons:

const COMPETITION_ID = 'YOUR_COMPETITION_ID'; // look up via /football/competitions?search=world%20cup
const today = '06-19';
const matches = await fetch(`/matches?competition_id=${COMPETITION_ID}&date_matches=${today}`).then((r) => r.json());

Then render "On June 19 at the World Cup..." with one match per year that matches.

Are penalty shootout results included?

Yes for every edition since 1982 (when the first shootout in a World Cup Final occurred — Spain 1982 SF actually). Each shootout returns shooter-by-shooter outcomes.

What data is available for the 2026 World Cup before it starts?

Full fixtures, venues, host cities, draw structure, and team metadata are available from the moment the draw is finalised. Live data (scores, lineups, xG) begins as matches kick off on June 11, 2026.