Goodbye sports reporter, hello sports reporter robot?
Students at Northwestern University are working on an algorithm that, if all goes as planned, will literally write sports stories without human intervention.
StatsMonkey, as the algorithm is called, is the creation of Medill students Nicholas Allen, Tian Huang, John Templon and computer science student Thu Cung as part of their work at the Interactive Innovation Project.
The invention takes raw sports data like box scores and play-by-play and fashions them into a readable journalism story.
But the machine doesn’t replace writers -- it supplements them.
“There are a tremendous number of sporting events that go uncovered, even at the college level,” Allen said to Medill Magazine, which originally reported on StatsMonkey. “So we think that there is great potential for our product in that space.”
“Our project is a way to keep journalism thriving,” Huang says. “By expanding coverage and freeing reporters from writing the basic game stories, we're giving journalists the time to really focus on the features and analysis that drive readers to their publications.”
To be sure, the program can’t plumb the depths of a slugger’s hitting slump, or report on a manager’s tirade on an umpire, but it can deliver the nuts and bolts of a story.
The StatsMonkey team gave their program story templates to work with (comeback win, blowout win, etc,) and the computer reads the data from the game and spits out some copy.
The applications of such a tool are many and varied -- imagine if your local paper covered every little-league game – but the depth of reporting is implicitly limited. That could change, however, with the next iteration of software. Soon enough, sports reporters could be obsolete.