Revealing Basketball's 10 Hidden Positions
For as long as basketball has been played, it’s been played with five positions. But through topological data visualization, Muthu Alagappan discovered the 10 positions hidden among them, with the power to help even the Charlotte Bobcats improve their lineup and win more games.
Jeff Beckham is a contributor to Wired's sports blog, Playbook, where we wrote about Alagappan's award-winning work. Alagappan is a basketball fan who works at Ayasdi, a data visualization company in Palo Alto founded by Stanford mathematicians. Ayasdi specializes in visualizing the shape of data, transforming huge amounts of information into beautiful interactive networks that instantly uncover hidden patterns, like which genetic markers best indicate a likelihood of ovarian cancer. It's called topological data analysis, and it can be applied to sports, too.
Come hear how Alagappan did it and what else he's uncovered in the data since his first discoveries.
Additional Supporting Materials
- How did you use data analysis to categorize players into 10 new positions, and what are those positions?
- How does this insight allow coaches, owners, general managers, and the everyday fan discover undervalued players, manage in-game decisions, optimize rosters, and draft more intelligently?
- Can this approach revolutionize basketball the way the "Moneyball" concept took baseball by storm? What are the similarities and differences between this analysis and the "Moneyball" way?
- Which combination of these positions do you think is most optimal (i.e. what is the best way to construct an NBA team using these positions)?
- How can topological data analysis can lead the way to more evolutions in thoughts in basketball and other sports?
Jeff Beckham, Contributor, Wired
Show me another