Home Tournaments Sabermetrics to Storytelling: Opening Up the Data in League of Legends Esports

Sabermetrics to Storytelling: Opening Up the Data in League of Legends Esports

by Graham Ashton

Mentioned in this article

As the head of esports insights for Riot Games, Doug Watson is responsible not only for numerically expressing the actions of players, but also measuring their impact on fans for media buyers. It follows an almost eight-year analytical background for agency Havas, working with clients including Fidelity, Air France, and Hyundai Kia, the latter of which is now a sponsor of League of Legends’ European league. “We aim to bring the voice of the fan together, and try to represent that to our property stakeholders,” he told The Esports Observer.

Watson explained that data and insights are used throughout the League of Legends esports ecosystem; from discovering which commercial partners to work with, determining where and events should be scheduled, to building out Riot’s digital product space. “What is the right way of creating the fan engagement that we’d like,” he said. “How we best create ways for them to follow the sport that we can?”

Having joined Riot Games in 2016, Watson has used the last few years to evaluate options when it comes to the publisher’s online digital presence. While China is a good example of a locally based region, with home arenas for teams and large-scale touring meet-and-greets with fans, other regions like the U.S. and Europe are more digitally based. 

Analytics have also helped Riot develop its pro-view product; a premium tool that allows viewers to watch esports matches from the perspective of a competitor, right down to individual mouse movements.

Credit: Riot Games

“It can also be a cohesion of trends in music to be brought into the scene, that can help create moments like K/DA,” he said. “That’s what we’re here to do; to represent that new voice and creating those experiences.”

In August of this year, Riot Games announced its first exclusive partner in the data space: Bayes Esports Solutions. The Berlin-based startup would not only deliver pre-match, real-time, and post-match insights from competitions, but would also have the right to develop its own products for LoL esports, such as match trackers. This is a project that had preceded even Watson’s four-year tenure at Riot.

“It’s taken a while because there’s a lot of complexity to that space,” he said. “There’s a lot of different ways it can be used, a lot of them can be very beneficial to our audience, and some of them are things we need to be very protective of, primarily the betting space.” 

Bayes is a joint venture between esports data tool developer DOJO Madness and sports data corporation Sportradar, the latter of which has also signed on as Riot’s integrity services partner for League of Legends—monitoring global betting activity for domestic and international competitions.

Credit: Riot Games

Watson further explained that by opening up its data to a broader market, the publisher can build tools for individual players to improve, and for fans to stay better connected to professional teams and players. “Think about what fantasy football did for the NFL a few years ago,” he said.  “The ecosystem can start to identify where are these opportunities and can help think beyond what we at Riot have a perspective of doing.” 

A number of the top teams in League of Legends are increasingly turning to traditional data companies to heighten their competitive performance. Team Liquid, after collaborating with SAP for over a year on the game Dota 2, would now use the software company’s data to aid in choosing champions and building strategies. Likewise, Cloud9 is working with Microsoft’s data suite to better adapt to patches—the fortnightly changes made to League of Legends that often completely rebalance the game.

“By enabling this to be accessible, we’re able to enable those partners to do the different types of sabermetric creation, and to look at it in a different way,” said Watson. “Historically our role in executing the esport is to be a storyteller. So the types of stats that we focus are things to enable us to tell better stories. To compare Doinb to Caps in a match-up.” 

“But if we enable the ecosystem to look at the data, there’s a lot of perspectives they have that we don’t. Some of those are from the teams in terms of performance, some are from agencies looking for new talent, and some are from fans looking to learn and get better at the game.”

Credit: Riot Games

Outside the realm of human gameplay, artificial intelligence labs are increasingly using esports games to further research in machine learning and open-ended learning systems. These include Dota 2’s OpenAI Five, created by the firm co-founded by Elon Musk, and AlphaStar, a StarCraft II AI that now ranks above 99.8% of human players. IBM has also adapted its AI software, used for data insights in the U.S. tennis open, to assist esports commentators.

“We’ve looked at how we can create better competition as well as better ways of training individuals, as well as to learn ourselves by leveraging AI in our development path,” said Watson. “I don’t think we have a definite plan in place yet for it, but it is something that we’re actively exploring.”

From a data perspective, Riot is already blessed with a game in the multiplayer online battle arena (MOBA) genre, one of the more complex types of games on the market. But for League of Legends’ tenth anniversary, the publisher announced its entry into the fighting game, card game, role-playing game, and first-person shooter genres, as well the mobile market—all of which carry their own considerations and possibilities for data collection and usage.

“Via League of Legends and its esports we’ve learned a lot as to what that data can enable, and what is the best way to bring it to market,” said Watson. “It’s led us to the APIs we’ve created for the game, and the ecosystem we’ve created around that.” 

Credit: Riot Games

Riot has already made a significant level of data available for the pre-launch of its card game, Legends of Runeterra. Through partners such as Twitch and Mobalytics, viewers were able to hover over and see the stats and abilities of individual cards on stream. 

Watson added that these insights would become valuable as Riot rolls out the new mobile spinoff League of Legends: Wild Rift, and evaluates its esports potential. “There are things that we would probably leverage, and may lead to cross-play promotion and scouting for new players, that might belong in the PC version. Again, it will depend on what we find is appropriate for the sport long-term.”

Source link

Related Articles

Leave a Comment