Pramana, a two-year-old company founded by former Intel execs, built SHIFT, which uses natural-language processing to simplify querying large datasets. Whereas an engineer might have been relied upon to look up the last time an American rookie scored in three consecutive home games to start his career, SHIFT offers an accessible web tool that can accomplish the same task.
“We vastly increased what we were able to do over the past season,” NHL digital production vice president Russ Levine said. During the playoffs, the league added a live stats blog to its media site, updated regularly in response to on-ice events. “That’s something we could not have done nearly as quickly without SHIFT,” Levine added.
The NHL was able to use the tool thanks to cleanup work done three years ago, ahead of its centennial season, to create a complete game database. The league became Pramana’s first SHIFT client in 2019, and weekly meetings between the sides have aided the product’s development with an eye now towards giving fans more access.
Levine says the tool has helped the NHL contribute to online conversation, and as the league starts doing more advanced tracking of its players, similar functionality could help educate fans. “In some ways hockey is the most chaotic sport there is,” he said. “It’s hard to articulate what makes a certain defenseman better than another just by watching. What are the things we can do statistically to help fans better understand what they’re seeing?”
Pramana co-founder Corey Patton said the company is in conversations with other leagues as well. “It’s a definite push for leagues across the landscape to have access to historical data,” he said. “Everything that happens in the sports world is done in the context of some historical data…. The power is knowing when something is uniquely relevant.”
Pramana is also looking to license SHIFT tech to consumer voice platforms and tech firms, among other potential clients. The company, which currently has 10 employees, has been self-funded so far.