SVG Sit-Down: Sportradar’s Brian Josephs Talks Peacock’s Powerful Performance View — and Just Telling Better Sports Stories

Sports-analytics firm’s product provides data-driven overlays to streaming service

Although Swiss sports-analytics giant Sportradar has built its brand based largely on providing the goods for the global betting industry, delivering quantifiable data and insights to the exploding business of live sports broadcast and streaming is now its fastest-growing revenue stream, with sales increasing 33% in the third quarter to nearly €44 million.

Sportradar’s Brian Josephs: “Through the AI models we’ve built on top of our player-tracking data, we have the ability to derive deeper analytics around the sports we cover. That’s not just for coaches and bookmakers but also for a broadcaster perspective.”

Among the hit products recently introduced in the U.S. by the 24-year-old company is Performance View for Peacock, which provides data-driven screen overlays that augment the fan experience for the streaming service’s recently commenced NBA coverage.

Sportradar is also being talked about for an innovative AI-driven tool it offered Amazon Prime Video, which lets Thursday Night Football announcers Al Michaels and Kirk Herbstreit predict oncoming blitzes better than the offensive coordinators themselves.

SVG recently caught up with Sportradar VP, North America, Brian Josephs to give us the download on what’s new with one of U.S, sports broadcasting’s most innovative vendors.

So what’s a nice Swiss multinational with deep roots in sports betting doing in a place like this?
We’ve actually been in the media business since 2015. That was right before I started. What Sportradar has done over the last decade is effectively build out a robust product portfolio and offering that help sports media, publishers, broadcasters, and now tech companies tell stories. This space continues to evolve. In a lot of ways, it mirrors what we do from a gaming standpoint.

In addition to sports-betting companies, such as FanDuel, and the pro leagues themselves, you also provide data to companies involved in broadcasting and streaming sports. What kind of data do you offer these companies?
We provide everything from your core player statistics to player tracking, betting, and odds. We also provide historical data. That’s one bucket. The second is advanced analytics. Through the AI models we’ve built on top of our player-tracking data, we have the ability to derive deeper analytics around the sports we cover, where we have that content available to analyze. That gives a whole other level of storytelling capability that’s not just for coaches and bookmakers but also for a broadcaster perspective. They like to look at individual matchups and lineup combinations and things like that as part of the way that they educate their audiences about the game and tell their stories.

What video providers do you work with?
We work with many of the national broadcasters and all the regional sports networks to provide the data, research, graphics, tech, and everything else that they use in their linear-broadcast production. We also launched a custom version of our 4Sight [AI-based graphics overlay tool], called Performance View, for NBCUniversal to use on Peacock for their NBA coverage. That provides an augmented streaming experience.

Your data also serves the actual players and coaches, too?
That’s correct. It’s for basketball and diamond sports, baseball and softball. We acquired a business called Synergy a couple of years ago. They specialize in data and video for player and team performance. And they have a very strong position with basketball and diamond sports, starting at the professional level globally all the way down through high school and amateur competition.

Drilling down to the high school level must require a small army, yes?
We have around 4,000 full-time employees. But, if we’re talking about data collection and the scouting network and all the data journalists involved, that number can balloon upwards to, I don’t know, 10,000? This year, we will collect data for over a million games or competitions globally, which is an absolutely wild number to say out loud.

We collect the data by a range of different methods, first and foremost being manual data entry. We have production groups and facilities around the world. Their facilities look like trading floors. Another data source comes from our partnerships with the leagues. We also use AI technologies to collect data. That’s something that we’ve invested in heavily to help improve the accuracy and quality of the data that we can collect.

We use a range of different methods to put it all together. We still have a tremendous research team, but where I think we’ve made a lot of strides recently is arming video-production folks with the right technologies to tell better and deeper stories and complement what they do with new technologies like AI.

Who’s Sportradar’s primary competitor?
Oh, goodness. Sometimes it feels like it’s every sports-technology company out there, in some form or fashion.

What’s your biggest product offering in North America?
That would be our broadcast-services offering. Again, everything we do is built on top of the data. Storytelling is a huge component of that. We have our Radar360 research application that is effectively a research tool where we house all our data that our clients can use to analyze and create different storylines in context. A lot of what you see in graphics or spoken on-air by the talent comes from Radar360. It arms on-air talent with the information they need to engage their fans.

Interactivity is becoming a sought-after product feature for sports-production tools, yes?
Features like watch-party and multi-language options are catching on. We are starting to see a shift to a bit of a user-controlled experience, where the viewer can choose their own adventure based on the content that they want to engage with. That has helped lead this shift — making sure the technology is available to do that but also to evolve from a data standpoint. People want to know the score, but now they also want to know, Why did that just happen? That’s where augmented streaming comes into play. That additional deeper storytelling insight based on the matchups on the court becomes far more relevant.

Any examples you can point to?
Thursday Night Football on Amazon Prime Video provides a good example. We have a tool that very accurately predicts blitzes based on where different defenders are lining up. That’s really interesting. And Peacock has this cool Player Postcard feature: whoever has the ball, the screen shows their shot chart and heat map, revealing where they shoot better or worse from. That’s part of the Performance View tool.

I was watching my USC Trojans lose a Big Ten game on FOX a few weeks ago. The announcers went on and on about how Illinois’s quarterback was coming off a bad week but still believes in himself and still believes in his teammates … blah, blah, blah. Do you tell those kinds of warm, fuzzy, annoying stories, too?
[Laughs] As much as I’m an Illinois fan who grew up in Chicago and I love [Illinois quarterback] Luke Altmyer, the answer is no. The stories we help broadcasters tell come from data, hard metrics, and insights. We’re not going to replace the sideline reporter anytime soon.

How did you get roped into this?
I was the second hire in New York when I started in 2016. Now we have at least a couple dozen people in New York. We’re certainly well into the hundreds here in North America. As for me, I’ve been in, call it, sales or client-facing jobs my entire career. I ended up at Sportradar through an old college professor who reached out to me seven years after I graduated and said, “I have something that combines sports fans and tech. Are you interested?” I’d never heard of Sportradar. That was almost 10 years ago.