SVG Sit-Down: Sportvision’s Adams Dishes on NHL Player Tracking, K-Zone on Each Pitch for ESPN
Sportvision has been extremely busy in 2015, playing a key role in ESPN’s deployment of K-Zone pitch tracking on every pitch for the first time and in the NHL’s experimentation with player-tracking technology at the All-Star Game in February.
From the 1st-and-10 line and the FoxTrax glowing puck to PITCHf/x and RACEf/x, Sportvision has been a virtual-graphics heavyweight for nearly two decades and is primed to play a key role in the emergence of virtual graphics and data visualization in the coming years. Last month, SVG sat down with Sportvision CEO Hank Adams to discuss ESPN’s deployment of K-Zone for every pitch, the NHL’s interest in player tracking, and where he sees the industry headed in the near and long term.
How important is ESPN’s use of K-Zone pitch tracking in the evolution of virtual graphics in live sports coverage?
This isn’t the first time we have done K-Zone live with ESPN; we have been experimenting with it off and on for years. But, this year, ESPN came to us and said they wanted to do it live and every pitch. There was a lot of back and forth to figure out what the look and feel should be. We wanted it to be distinctive but not overpowering. For example, the Alpha is slightly different inside the box. That’s one of the things that we worked with ESPN to come up with, and we will continue to experiment with it.
PITCHf/x on every pitch has actually been done before. For example, Turner in the playoffs will have a 2D graphic off to the side. But, in terms of the over-the-plate perspective, nobody’s ever done it for every pitch.
The pitcher/batter cat-and-mouse game is central to the sport of baseball. Highlighting that and showing how the pitcher is trying to work this batter is interesting and exciting and has lot of potential for helping the fans understand the game better. Baseball is one of those sports I’ve watched all my life, and I learn something new all the time about the strategy and the dynamic. So I think [this] has a lot of potential to explain what’s going on in the minds of some of the best pitchers and the best batters.
How has the fan reaction been so far?
I’ve seen people love it and others say it is distracting. By and large, I think it’s been positive. ESPN is a very sophisticated operation in terms of fan feedback, so they will be able to gauge that.
Do you expect the use PITCHf/x for every pitch to expand to more networks?
ESPN has K-Zone over-the-plate perspective exclusively, and we have a patent on it with them because they pioneered this concept a long, long time ago. That particular look and feel will be exclusive to ESPN, but we can do other in-perspective graphics: trails on balls, for example.
I have not seen anybody commit or propose that they would do that for every pitch. Over time, we have seen more and more use of the product as fans have demanded it. It’s really interesting to see the dynamic and have the visual there all the time. Personally, one of my favorites is when we see the whole at-bat with every pitch on a replay. Then you see the pattern of how they’re throwing the pitches. I think that’s great storytelling.
The NHL debuted player tracking using Sportvision technology at the All-Star Game in February. How do you see this system benefiting NHL broadcasts?
We did it on the skills competitions where we thought it made sense, like the breakaway. But where the technology is really well-suited to hockey is in-game.
For us, to make [virtual graphics] work as a broadcast effect, it has to be something that’s hard to see, happens a lot, and is important to the game. If you meet the criteria, you can enhance the broadcast in a way that is relevant for fans. It can’t be distracting or just highlighting the obvious.
For example, we used to measure home runs, but we realized that it happens maybe only twice a game, which isn’t very often. And a home run is a home run: it doesn’t count more if it goes farther. It seemed like we were doing something for the sake of eye candy, and it didn’t actually make an impact for fans.
With hockey, loads of things happen that you don’t see and appreciate when you watch it on TV. For instance, who’s on the ice? When they do shift changes, you generally don’t see who’s coming and who’s going because you’re following the puck. They will often dump [the puck] down in the zone, and, of course, the camera follows the puck. [But] everybody’s changing out, and you have no idea who is coming in and going out.
And [with] deflections into the goal, pucks up against the boards, or even moving along the boards from the game-camera point of view, you can’t see [the puck]. Yes, you can infer where it’s at, but it would be so much more interesting to actually be able to see it.
Being able to show the geometry on a penalty kill and how the defenders are positioning themselves is incredibly important. That’s what real hockey insiders talk about a lot, and we would be able to show that in a way that’s hard to do with those additional graphics. Since hockey is a game of flow, there is so much data and subtlety that none of us even know how the game works that we’ll be able to highlight.
Any update on when we might see this technology used again in live NHL coverage?
What we did at the All-Star game was the first [experiment]; we have not done it in an official game. But it’s in discussions now, and everybody is trying to figure out the right way to move forward on it. But my hope and expectation is that this rolls out soon. We received very good feedback, and the press was very positive about it.
With all four leagues embracing player-tracking technology, we seem to be entering a golden age of data-driven analysis and visualization for live sports coverage. How has Sportvision helped to bring that about, and what role will you play in the future?
I think it’s a different era these days in sports. I think everybody knows that you have to be at the forefront of the data revolution both from a coaching standpoint and from a fan-engagement standpoint. I think there is going be some really killer digital products possible with this, too.
For us, it’s been a seven-year overnight success. We have been tracking pitches for a long time, and we have been tracking NASCAR cars for a long time.
I’d say NASCAR was at the forefront early on. We were tracking every car live — mostly for broadcast storytelling — by putting arrows on the leader, who might be in the middle of a pack, and showing restarts and all that. And we would use that data for live broadcasts, but it would be stored in a flat file and [was] never looked at again. It was a couple years after we got going that we started thinking, “Wait a second, this could actually be interesting for a live digital product.” And so we came out with a product called RaceView and realized there were some really interesting stories that could be told with this. Things like that and PITCHf/x create a database that can be huge treasure trove.
Think if you could have what I call a complete digital record of an event. We’ve long thought of [the value of] a video record. I think that, long term, a digital record of an event is going to be just as valuable: if I get good enough data, I can not only tell you where a certain player was at a certain point and time or where the puck or football was, but I can actually render that play.
We’ve rendered cars, but they’re rigid objects, so it’s easy to do that. But I could, in the future, render a live football player, through a 3D rendering system, and I can put you on the field, I can put you in the press box, I can put you anywhere you want to be to interact with this event after the fact.
That is where I think [the digital record] is going to be incredibly valuable. I could hit the pitch live; I can try to be the goalie to stop the shot. I can transport myself into the event and interact with it. I think that is going to be revolutionary going forward.