FIFA World Cup 2026 Schedule & Lowest Ticket Prices (US & Canada)

Purpose

Return the full FIFA World Cup 2026 match schedule from SeatGeek, filtered to the US and Canada host cities only, as a JSON array where each object is { matchup, stage, date, time, venue, city, lowest_price }, sorted by lowest_price ascending. The entire 104-match dataset (teams, venue, city, country, kickoff datetime, and a live lowest ticket price per match) is server-rendered into the landing page's __NEXT_DATA__ blob, so this is a single HTTP fetch + JSON parse — no clicking, scrolling, or per-event navigation. Read-only; never adds to cart or checks out.

When to Use

Building a World Cup 2026 price tracker / "cheapest matches" leaderboard for US + Canada venues.
Pulling the complete fixture list (matchup, stage, date/time, venue, city) for the 91 non-Mexico matches in one shot.
Monitoring lowest_price drift per match over time (the value is live SeatGeek inventory).
Anywhere you'd otherwise scrape SeatGeek's infinite-scroll match list — the embedded JSON is faster, complete, and avoids the DataDome-protected UI entirely.

Workflow

Recommended method: fetch. One GET of the landing page returns all 104 matches inside __NEXT_DATA__. A residential proxy is required (SeatGeek fronts everything with DataDome — see Gotchas), but no browser, JS execution, or session state is needed.

Fetch the landing page through a residential proxy:
```
GET https://seatgeek.com/fifa-world-cup-tickets
```
With the browse CLI: browse cloud fetch "https://seatgeek.com/fifa-world-cup-tickets" --proxies (returns HTTP 200 with the full ~4MB HTML; a bare/un-proxied request gets a DataDome 403).
Extract the embedded JSON. Pull the __NEXT_DATA__ script tag and parse it:
```
<script id="__NEXT_DATA__" type="application/json">{ ... }</script>
```
The full schedule lives at props.pageProps.worldCupAllEvents2026 — an array of 104 SeatGeek event objects. (props.pageProps.allEvents is only the first paginated page of 25 — do not use it; use worldCupAllEvents2026 for the complete set.)
Map each event to the output shape:
- matchup — the two performers[] entries with type === "international_soccer" whose name does not start with "World Cup"; join their short_name (e.g. "Cape Verde", "Saudi Arabia") with " vs ". Knockout-round events carry no team performers — instead pull the bracket placeholder from title via regex -\s*([0-9A-Z]+ vs [0-9A-Z]+)\s*- (e.g. "2E vs 2I", "W97 vs W98"). Third Place and the Final have no placeholder at all → emit "TBD".
- stage — the performers[] entry whose name matches World Cup (Group Stage|Round of N|Quarterfinals|Semifinals|Third Place|Final), with the leading "World Cup " stripped → "Group Stage", "Round of 32", "Final", etc.
- date / time — datetime_local (local-to-venue), split into YYYY-MM-DD (chars 0–10) and HH:MM (chars 11–16).
- venue — venue.name. city — venue.city (the precise municipality, always present, e.g. "Inglewood", "East Rutherford").
- lowest_price — stats.lowest_price (a plain number in USD; equals the UI "From $X" price).
Filter to venue.country ∈ {"US", "Canada"} (drops the 13 Mexico matches). Sort by lowest_price ascending. Expect 91 rows (78 US + 13 Canada).

Browser fallback

If the proxied fetch is ever challenged, the identical JSON is reachable through a browser session — it's only ~30× more expensive, with the same parse:

browse open https://seatgeek.com/fifa-world-cup-tickets --remote using a stealth session created with --proxies --verified. The page itself loads without a DataDome interstitial (browse get title → "World Cup Tickets 2026 ... SeatGeek").
browse wait load then browse wait timeout 3000.
Read the blob directly — do not use browse snapshot (returns 0 a11y refs on this page) and do not use browse get value #__NEXT_DATA__ (a <script> has no value → returns ""):
```
browse eval "document.getElementById('__NEXT_DATA__').textContent"
```
Parse and map exactly as in steps 3–4 above.

Site-Specific Gotchas

DataDome on every entry point. The pre-run homepage probe returned 403 / datadome. A residential proxy (browse cloud fetch --proxies, or a --proxies --verified remote session) is mandatory; bare requests get a DataDome 403. The FIFA landing page itself does not present a captcha/interstitial once you're proxied — it renders straight to HTML with all data inline.
Use worldCupAllEvents2026, not allEvents. pageProps.allEvents is only the first 25 (paginated UI page); pageProps.worldCupAllEvents2026 is the complete 104. pageProps.totalEvents is 104 and is a good sanity check.
country is the literal string "US" (not "USA" / "United States"); Canada is "Canada", Mexico is "Mexico". Filtering on {"US","Canada"} yields exactly 91 of the 104 matches.
stats.lowest_price is live and is the displayed "From $X". It matches the UI to the dollar at fetch time and drifts a few dollars between fetches as inventory turns over (observed e.g. $223 vs $221, $774 vs $804 minutes apart). It can be null if a match momentarily has zero listings — treat null as "no price" and sort it last. All 91 had a price during testing (range ~$171 group-stage to ~$7,869 Final).
city vs host-city label. venue.city is the precise municipality (Inglewood, East Rutherford, Foxborough, Santa Clara, Arlington, Miami Gardens…). The FIFA marketing host-city name (Los Angeles, New York, Boston, San Francisco, Dallas, Miami…) is in venue.marquee_city, but that field is null for roughly half the venues (Toronto, Vancouver, Houston, Philadelphia, Atlanta, Seattle…), so prefer venue.city for a field that's always populated.
Knockout matchups are placeholders, not teams. Round-of-32 through Semifinals encode bracket slots in the title ("2E vs 2I", "W97 vs W98"); the Third-Place match and the Final have no placeholder. These are genuinely TBD until the bracket fills — emit the placeholder where present, else "TBD".
Don't drive the visible list. The on-page match list is an infinite-scroll React component; scraping it row-by-row needs scrolling + reflows and gives strictly less data than the JSON. browse snapshot returns 0 refs on this page. Always read __NEXT_DATA__.
Lighter JSON endpoint exists but isn't standalone-stable. GET /_next/data/{buildId}/performer-tickets.json?slug=fifa-world-cup returns the same pageProps as a 2MB JSON document (no HTML). Confirmed working, but buildId rotates on every SeatGeek deploy and must first be read from a live page's __NEXT_DATA__ (buildId field), so it's only a useful shortcut once you already hold a fresh buildId — otherwise just parse the page HTML.

Expected Output

A JSON array sorted by lowest_price ascending. 91 objects (78 US + 13 Canada). Shapes:

[
  {
    "matchup": "Cape Verde vs Saudi Arabia",
    "stage": "Group Stage",
    "date": "2026-06-26",
    "time": "19:00",
    "venue": "Reliant Stadium",
    "city": "Houston",
    "lowest_price": 171
  },
  {
    "matchup": "2E vs 2I",
    "stage": "Round of 32",
    "date": "2026-06-30",
    "time": "12:00",
    "venue": "AT&T Stadium",
    "city": "Arlington",
    "lowest_price": 552
  },
  {
    "matchup": "W97 vs W98",
    "stage": "Semifinals",
    "date": "2026-07-14",
    "time": "15:00",
    "venue": "AT&T Stadium",
    "city": "Arlington",
    "lowest_price": 2510
  },
  {
    "matchup": "TBD",
    "stage": "Final",
    "date": "2026-07-19",
    "time": "15:00",
    "venue": "MetLife Stadium",
    "city": "East Rutherford",
    "lowest_price": 7869
  }
]

Field notes:

matchup — "<Team A> vs <Team B>" for group stage (short names); bracket placeholder ("2E vs 2I", "W97 vs W98") for Round of 32 → Semifinals; "TBD" for Third Place and Final.
stage — one of "Group Stage", "Round of 32", "Round of 16", "Quarterfinals", "Semifinals", "Third Place", "Final".
date YYYY-MM-DD, time HH:MM (24h, local to venue).
lowest_price — number (USD); may be null if a match has no active listings (sort last).

find-fifa-tickets