Nieman Foundation at Harvard
HOME
          
LATEST STORY
What journalists and independent creators can learn from each other
ABOUT                    SUBSCRIBE
March 5, 2014, 1:55 p.m.
LINK: kpq.github.io  ➚   |   Posted by: Joseph Lichterman   |   March 5, 2014

NFL teams spend the offseason reflecting on the past season and preparing for the next one. So like any good playcaller, the team at The New York Times’ 4th Down Bot is spending the months between the Super Bowl and the start of the 2014 preseason examining what it’s learned and how it can improve for the coming season.

The 4th Down Bot performs realtime analysis of every fourth down play in the NFL and determines whether a team should go for the first down, kick a field goal, or punt the ball away. The bot was a collaboration between the Times and Brian Burke of Advanced NFL Stats, who originally built the code.

On his personal GitHub today, Kevin Quealy, a graphics editor at the Times, writes about how the project was developed and outlined some of the bot’s successes (more than 10,000 Twitter followers!) and areas where it could be improved (a slow response time). The bot launched in time for Week 13 of the NFL season (in the midst of my Lions’ annual late season collapse, when the bot was a welcome tool to further question now-fired coach Jim Schwartz’s competence). Quealy said the bot probably should’ve launched sooner — a non-football kind of MVP — and he outlines a few other issues he might try to address in training camp:

It could feel more “live”. The lag between the end of the play and the analysis takes about a minute, but sometimes the delay on the play-by-play data lagged a bit, which meant you were getting bot analysis well after the other team started its drive. This isn’t ideal, but there just wasn’t much we could do about it.

Because it was programmed to analyze decisions that already happened, some aspects of N.F.L. play aren’t captured well. For example, when a team intentionally takes a penalty on 4th and 1 near midfield, the bot applauds the punt on 4th and 6 without properly scolding the 4th and 1. This particularly annoyed Aaron Schatz of Football Outsiders, who later got over it.

As many statisticians noted, it could display uncertainty better than it does. From my perspective, that’s the most legitimate criticism, and we hope to improve on it next year.

In his post, Quealy wrote that they’re “hoping to introduce a cousin or two this summer, too.” In a followup email, Quealy told me he’s looking to apply some of the successful aspects of the 4th Down Bot to other fields — possibly baseball or politics.

“But it’s not like I have a super secret project on my desktop that is already [4th Down Bot’s] next of kin,” he wrote. “Plus, we’re hoping to add features and build on our audience with 4th Down Bot, which is not an insignificant amount of work.”

Show tags
 
Join the 60,000 who get the freshest future-of-journalism news in our daily email.
What journalists and independent creators can learn from each other
“The question is not about the topics but how you approach the topics.”
Deepfake detection improves when using algorithms that are more aware of demographic diversity
“Our research addresses deepfake detection algorithms’ fairness, rather than just attempting to balance the data. It offers a new approach to algorithm design that considers demographic fairness as a core aspect.”
What it takes to run a metro newspaper in the digital era, according to four top editors
“People will pay you to make their lives easier, even when it comes to telling them which burrito to eat.”