The graphics provides fans with a deeper explanation of what’s happening on the field of play
AWS and Oval Insights have unveiled a series of data-driven graphic insights they have brought to rugby, as a result of having built machine learning models utilising data from every televised rugby union game to date.
The machine learning with AWS has enabled Oval to model win predictors and player ratings, and showcase great player performance, including for players who are often hidden from view.
Oval has a data collection center in Leeds that collects data on every action that takes place on the pitch for every elite Rugby Union match in the world.
It uses a combination of expert, human analysts and AI models to pull in live video feeds from every televised game and capture key metrics such as touches, carries, kicks and tackles.
Oval analyses around 3,000 individual data points for every match to provide the raw material for its machine learning models.
The company then uses AWS to store this data in a workable form, and has built an ensemble of bespoke machine learning models on AWS, which works as a pipeline with the output from one model feeding into the next.
All of this analysis runs on AWS, using Elastic Container Services (ECS) Instances. In this way, Oval generates more than 800 metrics for use by its customers.
Oval uses AWS Lambda to deliver real-time predictive insights to broadcasters live, via APIs. The APIs include in-game win predictors and success likelihood for kicks.
The different environments in which these insights are used include clubs and national sides, broadcasters and media, fantasy leagues and fan engagement on social media.
Oval Insights is an official data provider to the Premiership, United Rugby Championship, European Cup, Sky Sports and Channel 9 Wide World of Sports.
All the on-screen data in TNT Sport’s Rugby Premiership coverage comes from Oval.
It includes in-game win predictors, predictions of the chances a kick will be successful, and a plethora of insights used by commentators and pundits. This includes details such as what happened when the two teams met in the past, or which set of forwards usually dominates in the scrum.
Oval also creates a Player Impact Score, which is generated from a very complicated data set, and provides a score out of 100 that’s an easy way to compare and assess players’ contributions. Players get a ranking for kicking, passing, catching, carrying, and so on.
No comments yet