SVG Summit: With the Right Tools, Big Data Yields Big Results for Broadcasters
Gone are the days when the intricacies of sports could be described using a few words and a replay. Much to every stats geek’s delight, sports broadcasts are now replete with the analytical tools to mine and translate piles of raw data into digestible on-air graphics. The question is, how are broadcasters tackling such a daunting task? More importantly, how much is too much?
“When we’re looking at data, you’ve got two questions that you ask,” says Brian Perkins, director, IT production systems, MLB Network. “One is that, is it important? Does it tell you about what just happened, what is happening, or what’s about to happen? And then for a broadcaster, you add in, does it make good TV? We’ve been predicting who’s going to win in a particular pitcher-batter matchup for forever; now, we would just be doing it with more education and more numbers on our side.”
At last week’s SVG Summit, executives from the leading manufacturers of graphics tools convened to discuss the concept of big data. “When you talk about big data,” says Dr. Stephan Würmlin Stadler, EVP, Sports, Vizrt, “I think it is everything that is not straightforward to visualize.”
Echoes ChyronHego CTO Sören Kjellin, “As soon as you have enough data so that it’s not possible to review it, then I would say that its big data… We’re doing these player-tracking things where you gather three to four million data points per match, and that is obviously completely impossible to review. That becomes big data.”
Player-tracking tools, like ChyronHego’s TRACAB system, have become a staple of NFL broadcasts, enabling networks to quickly identify athletes on screen and follow them throughout complex plays.
“[We] look at whatever technology makes sense for that sport,” says Ryan Zander, SVP/GM, baseball and motorsports, Sportvision. “In some cases, that’s instrumenting a racecar with a GPS system [or] putting cameras in a Major League Baseball park to capture pitches and hits and then working through the workflow of that data until we can create a presentation that makes sense for our broadcast partner, a consumer with an online application, [and] a lot of cases, with leagues, front offices, and teams.”
Motorsport and sailing broadcasters have also turned to tracking technologies, both to identify participants and make a complex sport easier for the casual viewer to follow.
“We’re all about game data, what goes on on the field of play,” says Brian Kopp, SVP, sports solutions, STATS. “Because we’re always about data and have been since we were founded, it’s not a challenge for us. I think it’s translating that into the story telling [and] working with our partners both on the media side and the team side. That’s what we’re all about.”
The panelists agreed that, when approaching big data, it’s all about matching the right tools to the right sports. Or, in the case of Bloomberg Sports, leveraging an infrastructure already in place.
“At Bloomberg, the company’s been dealing with enormous volumes of data for 30 years in the financial world, aggregating tremendous streams of data from vast arrays of sources,” says Bill Squadron, president, Bloomberg Sports.
“Our approach to the notion of big data is really consistent with the way Bloomberg has always approached it, which is to say that the volume of data is not really the issue. The issue is, how do you make sense of it? How do you create value out of it? How do you find functionality that makes sense for people?
“What we’ve tried to do in sports for the past four or five years is look at any data source, as long as its accurate and reliable, and try to figure out ways to create value for potential customers whether they’re fans or broadcasters or leagues or teams or others,” he continues. “That’s really the way we’ve approached the concept.”