Sportswriting – so easy a computer could do it

by Aaron Selbig
It’s been said that if you put a thousand monkeys in a room with a thousand typewriters, one of them would eventually write “Hamlet.”
But could they come up with a good sports story?
That line of thinking – one that inherently equates sportswriters with monkeys – is presumably what led a small of group of eggheads at Northwestern University to come up with Stats Monkey, a recently unveiled computer program that automatically generates sports stories based on nothing more than the stats from a game.
Basically, Stats Monkey takes the box score and play-by-play of a football, baseball or hockey game and analyzes which plays and/or players stood out and what the dominant narrative arc of the story should be. It then plugs in all the filler – key stat-lines and quotes from the coaches and/or players – and writes a story based on the “inverted pyramid” style of journalism – where the most important information appears first.
My first thought upon hearing about this fascinating new technology was “Uh oh. If this thing really works, I might soon be out of a job.” My second thought – which came about a nanosecond later – was “Man, if I had something like this, I could sleep ‘til noon everyday.”
I can identify with the Northwestern eggheads – and probably most readers of sports journalism – who think of sportswriters as mail-it-in hacks who crank out story after uninspired, cliche-laden story about the latest football, baseball or hockey game. I’ll be the first to admit it – most sportswriting is crap.
Some of it, however, is awesome. If it’s done right, a good sports story about a compelling game can be the most vivid, colorful thing in an entire newspaper.
And it’s a really, really difficult thing to do. Writing game stories about team sports, as a matter of fact, is the toughest thing to do in all of journalism. After all, how many different ways are there to say a baseball team “brought its A-game?” Or that the football team is “taking things one game at a time?” Or that the underdog hockey team “snatched victory from the jaws of defeat?”
And what about individual racing sports like swimming or cross country skiing? How many different ways are there to describe a foot race?

NEW YORK – At the New York Marathon today, some people ran faster than others.

Stats Monkey’s prefab baseball stories are only slightly more compelling. Here’s a sample story, based on statistics from an Oct. 11 playoff game between the Boston Red Sox and the Los Angeles Angels:

BOSTON — Things looked bleak for the Angels when they trailed by two runs in the ninth inning, but Los Angeles recovered thanks to a key single from Vladimir Guerrero to pull out a 7-6 victory over the Boston Red Sox at Fenway Park on Sunday.
Guerrero drove in two Angels runners. He went 2-4 at the plate.
“When it comes down to honoring Nick Adenhart, and what happened in April in Anaheim, yes, it probably was the biggest hit (of my career),” Guerrero said. “Because I’m dedicating that to a former teammate, a guy that passed away.”
The Angels clinched the AL Division Series 3-0.

Certainly there are human-generated baseball stories out there that are worse than this one – two or three of them may even be mine. And the Stats Monkey’s story does seem to have all the basic, bare-bones information that a casual baseball fan would be looking for – who won, who played especially well and how the game played out.
But it’s missing one critical thing that makes or breaks any piece of writing – the human conveyance of human emotion that can only be provided by, well – a human.
So – for the time being, at least – I’m not going to worry about computer programs like Stats Monkey edging me out of a job. I’m going to keep plugging away – probably with varying degrees of success – trying to extract the human components out of sports stories and make them interesting for everyone.
And if that doesn’t work, I’m hiring me some monkeys.

Contact the writer
Posted by on Oct 28th, 2009 and filed under Extra Innings. You can follow any responses to this entry through the RSS 2.0. Both comments and pings are currently closed.

1 Response for “Sportswriting – so easy a computer could do it”

Comments are closed

Like us on Facebook