Week Four Activity

Recent innovations in data collection and analysis have transformed the quality and quantity of performance metrics available to all involved in sports, whether at the elite or amateur level. From the comparison workout reports posted with recreational tracking applications such as RunKeeper and Endomondo, to the live tracking and post event analysis of skiers, sailors, marathon runners and participants in many other sports, coaches, athletes and enthusiasts are able to monitor and record activities at a level of detail previously unimaginable just 10 years ago.  

Much of that data has a spatial or locational component—the position of the ski slalom gates, the distance covered by an endurance runner, the length of the pitch, the height of the dive board and so on. Sports are also by necessity confined to a specific location, or place, such as a tennis court, football pitch or cross country track (Clark and Kerski, 2014).  

The practice of data mining is being increasingly applied to the analysis of sports data with the explicit aim of finding innovative ways to improve performance, or motivate participants. Data mining is generally described as the analytical process of discovering patterns and relationships and a 'search for new, valuable, and non trivial information in large volumes of data' (Kantardzic 2011, p.2). Retired American football player Chris Kluwe recently commented 'Now your IT department is just as important as your scouting department. Data mining is not for nerds any more. It’s also for jocks.' (Wohlsen, 2014)

The development of these analytical techniques has helped to improve not just performance, but also produced better equipment and clothing, safety for participants, access for spectators and general awareness and interest in physical activity. They also have the potential to provide invaluable behavioural insights for many interested parties, including marketing and insurance companies, government agencies and strategic health authorities. 

With so much of this information now being made available online, which includes specific references to locations, geotagged photos and GPS tracks, just who has access to the data and what else can they do with it? Is everyone familiar with the privacy settings for their chosen publishing platform? As with all technological innovations, there are positive and negative aspects. It’s not usually the data that are the problem, it’s what others choose to do with them that can be the issue. 

The following short video and articles describe the results of some of the analysis of personal location data collected over a period of time in the USA.

 Quantified Self Public Health Symposium - The analysis of personal tracking data for improved public health awareness (8:32mins)   

 Fitness Tracker Data Shows Who the Bay Area Earthquake Woke Up - Profiling fitness enthusiasts while they slept 

 The countries that went to bed early on NYE 2014 (another example from Jawbone)

 Do you know where you'll be 285 days from now at 2pm? - A thought provoking article on predicting future actions (not strictly sport related but relevant)


Having read the articles and watched the video, now have a go at data mining yourself in this week's activity.  

Refs:

Clark, J and Kerski J. (2014). Using Geotechnology Tools in Sports Coaching. In Practical Sports Coaching (Routledge). In Print.

Kantardzic, M. (2011). (2nd ed). Data-Mining Concepts. In Data Mining: Concepts, Models, Methods, and Algorithms (John Wiley and Sons, Inc., Hoboken, NJ, USA): pp. 1-24. 

Wohlsen, M. (2014). Augmented Reality Is About to Turn Football Into a Real-Life Videogame. Retrieved: 27 Dec 2014. http://www.wired.com/business/2014/03/future-winning-super-bowl-department/?mbid=social_twitter

Last modified: Thursday, 1 January 2015, 1:17 PM