• Screens are getting smaller, attention spans shorter • And creating sports content is expensive • Today we built a programmatic create engaging sports content • If you want to see the highlights of that basketball game you missed, you can build you a the story of the game in a few key snapshots without any human intervention
How it works
Our system compares play-by-play data, player statistics, and the games in-play odds to mark key, which could be times when the likelihood of a team's victory changes significantly, or a historically rare occurrence. We use timestamps and other meta data to associate it with text, pictures and video, and generate our slides.
We are using python to scrap play-to-play data and team information from the respective sports sites. Also, we are using node.js to drive the operations of the platform.
Challenges I ran into
The flow process was complicated
Accomplishments that I'm proud of
Our team managed to put the platform together as team despite the differences in skill sets
What I learned
Teamwork is the key to success
What's next for Game Recap Robot
We hope to build and include heatmapping as well as our own predictive models into our platform