Pandascore turns e-sports matches into sets of data that can instantly be manipulated or archived for later. When you watch a basketball game on TV, the screen is filled with stats while the game is still going on. And if you’re missing a match, you can also access all this data from a browser on the NBA’s website or any sports website. But this isn’t true for a League of Legends game today. That’s what Pandascore wants to fix.
The French startup uses a standard live stream on Twitch or YouTube to figure out what’s happening right now. Stats are automatically compiled during the game. Conversely, during a basketball game, human beings with actual brains press buttons to register rebounds and three-pointers.
Pandascore takes screenshots and analyzes information on the screen. You can figure out quite a lot of things by just having the algorithm look at the mini map in the corner, the scoreboard and the kill feed. But the startup made sure that Pandascore could remain versatile. Even if the interface moves around, Pandascore automatically adjusts its detection zones.
In traditional sports, the standard is the human eye. The startup wants to provide at least the same level of accuracy. “We don’t use a module if it’s not better than the human eye,” co-founder and CEO Flavien Guillocheau told me.
You can automatically adjust the refresh rate so that Pandascore pulls information 10 times per second or once every few seconds depending on the importance of the match. This data is then analyzed, sorted and turned into an API. Developers can then take advantage of this API to display this data anywhere they want and more.
Pandascore currently works with League of Legends, FIFA, Dota 2 and soon Overwatch. The company has been working with e-sports teams, such as PSG’s League of Legends and FIFA teams. They get reports and insights to improve their skills. Competitions have also tried using Pandascore, such as the DreamHack in France or the Gamers Assembly.
But the startup slowly discovered that this kind of product made sense for betting websites. What if you could bet on the player who is going to destroy the most turrets in a League of Legends match? Pandascore hopes that betting sites in the U.K. will rely on its product to create new kinds of bets.