WorldRowing com: End To End Data Analysis
summary
WorldRowing.com data gets scraped with Python using requests and BeautifulSoup into a reproducible pipeline that required heavy cleaning because of missing IDs and malformed fields; roughly 50,000 athlete pages yield about 3,300 Olympians but bios contain absurd errors (dozens of heights >3 m and <1 m). Analysis uses NetworkX to build team graphs and weighted synergy ideas; the scraper and cleaning work well enough to reveal trends but data quality limits confident causal claims.
AI-written summary · about
chapters (11)
- 0:00WorldRowing data project intro
- 1:28Why analyze rowing performance
- 3:15Competitive rowing basics
- 7:53How we collected the rowing data
- 10:42Cleaning rowing competition data
- 15:08Exploring rowing trends
- 17:56Athlete success and junior development
- 21:12Key findings and next steps
- 24:51Athlete collaboration networks
- 29:24Future challenges
- 34:00Q&A
related talks
21mVisualising Enterprise Data with D3
22mVisualising Enterprise Data with Angular & d3
43mData Sketches | A year of exotic data visualizations
23mSport analysis with Python
40mOpen Data Dashboards & Python Web Scraping
53mThe Science of Showing Up [Opening Key note]
29mMissing the Apex: Predicting F1 Races with Python
41mEnhancing Event Analysis at Scale: Leveraging Tracking Data in Sports