Major League Baseball and Amazon Web Services announced an extended partnership earlier this week to leverage Amazon’s cloud-based machine learning technology to enrich the statistical data presented in baseball broadcasts.
The two companies have partnered since 2014, but this broadened working relationship will see AWS automate and digitize much of MLB’s scorekeeping and continue building on the potential of the Statcast tracking data. Amazon’s SageMaker will be used to assist baseball in a variety of tasks. One of which will be to build a pitch-prediction tool that will account for the pitcher, batter, catcher, and situation to forecast the type and location of the next pitch. The various metrics produced will be made available at ballparks, on MLB digital properties, and on in-game broadcasts.
“Incorporating machine learning into our systems and practices is a great way to take understanding of the game to a whole new level for our fans and the 30 clubs,” said Major League Baseball CTO Jason Gaedtke in a press release. “We chose AWS because of their strength, depth, and proven expertise in delivering machine learning services and are looking forward to working with the Amazon ML Solutions Lab on a number of exciting projects, including detecting and automating key events, as well as creating new opportunities to share never-before-seen metrics.”
In conjunction with natural language processing service Amazon Comprehend, SageMaker’s AI capabilities will also try to produce live analysis in the style of iconic baseball broadcasters.
“MLB has been collecting statistical data on its players and clubs for decades, and turned to AWS in 2015 to take its game-day stats to the next level so that fans can dig deeper into advanced metrics that ultimately enhance enjoyment of the game,” Mike Clayville, AWS’ Worldwide Commercial Sales VP, said in the release. “AWS has the broadest and deepest portfolio of cloud services with the best security and proven operational expertise, which is why MLB chose to work with the world’s leading cloud to build, run, and enhance Statcast.
“We are excited to work with MLB as they continue their journey in the cloud, leveraging AWS machine learning services to capture previously immeasurable aspects of the game and find new ways to enhance the fan engagement model for baseball.”
SportTechie Takeaway
Cloud computing is estimated to be an $82 billion industry in 2018, Reuters reported, with AWS accounting for a third of the market in the year’s first quarter. This increased partnership should continue to enhance its profile while MLB fans are likely to benefit from new insights into the game. Statcast produces roughly seven terabytes of data per game and 17 petabytes annually—far more than any human-centric analysis can ever begin to digest properly. Harnessing the raw computing power of AWS can help to unlock more meaningful observations and conclusions.