Sports scientists now have the ability to gather vast amounts of data on players and the action while games are in progress, but they are still struggling to process the data in meaningful ways.
New research from the University of Sydney's Joachim Gudmundsson and Michael Horton addresses the challenges sports scientists face in gaining meaningful insight from data. For example, one of the most significant tasks involves understanding how players can dominate parts of the field or court near them, which requires sports scientists to take into account the players' momentum, work out whether they are open to receive a pass, and measure the pressure from others closing down the space around them. Sports scientists would have to find a way to incorporate such factors into models, but perfecting algorithms is only half the battle.
Another class of problems comes from analyzing game-play data. One puzzle asks whether it is possible to determine team formation or the type of marking used by the defensive team, given the list of player trajectories and event logs for a period during the game.
Gudmundsson and Horton also offer ideas for crunching the data effectively, and the next stage of the rapidly evolving field is to help improve player performance and provide teams with a competitive edge.
From Technology Review
View Full Article
Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA
No entries found