Tour de France Rankings & Stages Scraper

Scrapes official Tour de France jersey rankings (GC, points, mountains, youth, and team) plus stage metadata (start/finish cities, distance, elevation, stage type) directly from letour.fr. Covers all 21 stages — updated daily during the July race window.

What does it do?

Each run retrieves:

• Stage metadata — start/finish cities, total distance, elevation gain, and stage classification (Flat, Hilly, Mountain, Individual Time Trial, Team Time Trial)
• Jersey rankings for each stage across up to 7 classification types: General (GC yellow), Points (green), Mountains (polka-dot), Youth (white), Team, Combative, and Lanterne Rouge
• Rider details — position, bib number, name, country (ISO 3-letter code), team name, accumulated time, time gap to leader, and points

Data is sourced from the official site's AJAX ranking endpoints, which return server-rendered HTML for every completed stage.

Why use this instead of scraping ProCyclingStats?

ProCyclingStats is the de facto third-party cycling data source but it is protected by Cloudflare (verified 403 in pre-flight testing). letour.fr serves the same official rankings from clean, datacenter-friendly HTML with no bot protection — no residential proxy required.

Input

Field	Type	Default	Description
year	integer	2026	Tour de France year to scrape
stageNumbers	array of integers	all 21 stages	Filter to specific stages, e.g. [1, 2, 21]
rankingTypes	array of strings	all types	Which jersey rankings to scrape: general, points, mountains, youth, team, combative, lanterne_rouge
maxItems	integer	0 (unlimited)	Hard cap on total records saved

Output dataset fields

Field	Type	Description
year	integer	Tour de France year
stage_number	integer	Stage number (1–21)
stage_type	string	Flat / Hilly / Mountain / Individual Time Trial / Team Time Trial
start_city	string	Stage start city
finish_city	string	Stage finish city
distance_km	number	Stage distance in km
elevation_gain_m	integer	Total elevation gain in metres
ranking_type	string	general / points / mountains / youth / team / combative / lanterne_rouge
position	integer	Rider position in the ranking
bib	integer	Rider bib number
rider_name	string	Rider abbreviated name (e.g. "T. POGACAR")
rider_country	string	Country code (ISO 3166-1 alpha-3, e.g. "SLO")
team_name	string	Full team name
team_code	string	Team 3-letter code (where available)
finish_time	string	Accumulated race time for general/youth/team rankings
time_gap	string	Time gap to race leader
points	integer	Points total for green/mountains rankings
source	string	Source URL of the ranking data

Sample output (Stage 21 GC, 2025)

{
  "year": 2026,
  "stage_number": 21,
  "stage_type": "Flat",
  "start_city": "Thoiry",
  "finish_city": "Paris Champs-Élysées",
  "distance_km": 133,
  "elevation_gain_m": 1000,
  "ranking_type": "general",
  "position": 1,
  "bib": 1,
  "rider_name": "T. POGACAR",
  "rider_country": "SLO",
  "team_name": "UAE TEAM EMIRATES XRG",
  "team_code": null,
  "finish_time": "76h 00' 32''",
  "time_gap": null,
  "points": null,
  "source": "https://www.letour.fr/en/ajax/ranking/21/itg/2b9.../subtab"
}

Usage tips

• During the race window (July 4–26, 2026): Run daily to capture updated standings after each stage finishes. Stage 1–20 rankings update that evening; Stage 21 (Paris) closes the tour.
• Historical data: Stage results from the 2025 Tour are accessible at the same URL paths.
• Future stages: Ranking tabs are only populated for completed stages. Running the scraper on a future stage returns 0 results for that stage — no error is thrown.
• Scaling: For a full run (21 stages × 5 jersey types = ~105 ranking tables × ~160 riders = ~16,800 rows), a single run completes in under 3 minutes at the default concurrency of 5.

Technical notes

The actor uses a 4-level hierarchical crawl:

1. /en/overall-route → discovers all stage links
2. /en/stage-N → extracts stage metadata (city names from page title, distance/type/elevation from stageHeader__infos blocks)
3. /en/rankings/stage-N → discovers AJAX subtab URLs (hash-keyed, rotated per year)
4. /en/ajax/ranking/N/<type>/<hash>/subtab → parses rider rows from the rankings table HTML