During The Championships, IBM typically captures 3.2 million data points from 19 courts across 13... [+]
It’s ‘Wimbledon week’ here in the United Kingdom. Actually, it lasts two weeks, but people like to say it that way. If you drive or walk past the so-named All England Club Wimbledon outside of the two-week period of the championships it appears mostly dead. Watch Wimbledon by numbers on Forbes to learn that there are in fact only 375 full permitted memberships of the club itself.
Deeper sports analytics sense aggressive play
But this week Wimbledon is full of life, Pimms and an almost unfeasible amount of rather expensive strawberries and cream. As the tennis gets underway today we can once again see the ‘Hawkeye’ ball-tracking technology system being overlaid with core game analysis data from IBM.
The combination of the two systems means that computer-based big data analytics is being used to track not just whether shots are in or out; the software ‘knows’ enough about both previous and current player behavior to then track speed and ball bounce shot placement in any game to determine whether the player in being more (or less) aggressive than their normal style of play.
Of course it would be unfair to offer this kind of insight to a competitor during the normal course of play, but for fans (and for coaches afterwards) this kind of insight suggests just how digitized our tracking of sports performance is becoming.
New this year is the Slamtracker ‘Keys to the Match’ feature, built on IBM's predictive analytics technology (SPSS). This mines over eight years of Grand Slam Tennis data (~41 million data points) to determine patterns and styles for players when they win.
“Prior to each match, the IBM Keys to the Match system runs an analysis of both competitors' historical head-to-head match-ups, as well as statistics against comparable player styles to determine what the data indicates each player must do to do well in the match,” details the firm.
Tennis pros operate the computers
As noted previously, deep in the IBM Wimbledon bunker lives a group of data scientists and techies. Despite this geek team being in place, the job of hard-core data entry (to track the tennis being played) is carried out by tennis professionals. Turns out it’s easier to train tennis professionals to use the software system than it is to train a techie to really understand professional tennis in quite the same way.
So what’s actually new in 2016?
In terms of new toys and developments this year, IBM has launched an Apple TV app for Wimbledon with real-time scores, a Live @ Wimbledon studio show and three live radio channels with supporting video and photo content. The core app itself has also been updated for iOS and Android.
Also new this year is a Cognitive Command Center built using IBM’s Watson and hybrid cloud technologies. This technology ingests feeds across multiple social media channels to automatically understand, reason and learn the most relevant and emerging topics of conversation as they relate to Wimbledon.
“By identifying common topics of interest, IBM can help Wimbledon identify opportunities to better serve related articles, posts and images. For example, the Cognitive Command Center could identify emerging conversations around a Swiss football (soccer) game at the same time as a peak in interest around a moment in a Roger Federer match. Using these insights, Wimbledon will be able to make rapid content decisions to engage and inform sports fans during a summer filled with numerous major sporting events,” explains Sam Seddon, IBM Wimbledon client executive at IBM UK.
American football and basketball analysis also
It goes even further (and before you ask, yes, these same technologies are being applied in American Football, Basketball and more), IBM Watson’s machine learning engine can also be directed to sentiment being expressed on social media enriched with Natural Language Processing (NLP)… and then do that in real time and then feed that back through the IBM Cognitive Command Center and you can see how much the whole mood and feel of the event is being digitized.
IBM claims to be able to do all this and be able to perform calculations in a sub-second response time and 100% accuracy.
A previous previous Forbes story on this subject rightly called out tennis for being a ‘fragmented decentralized landscape’ due to the autonomous nature of the events and the difficulties associated with aligning a uniform data tracking system -- and yes, that issue still pervades to a degree.
While these issues still need tabling, one could reasonable argue that the more sports events we see (not just tennis) using digital tracking and machine learning technology, the more we will be able to refine our approach to this subject.
All this technology and there’s still no WiFi at the Wimbledon club ground -- well, they wouldn’t want to spoil the tennis with people playing around on the web now would they?
