The World Cup kicked off yesterday with a fantastic (albeit controversial) game between Brazil and Croatia, with Brazil winning 3 – 1. Over the next month, we soccer fans will be glued to our screens—tablets, phones, TVs, you name it. We’ll be watching games, replaying highlights, reading articles, checking stats, tweeting, messaging our friends. The World Cup will consume our lives.
As a soccer fan, I started worrying that this would result in a significant drop in my productivity, which, you know, wouldn’t be that great since we here at Chartbeat are constantly working to deliver fantastic products and provide key data insights. I got to thinking about how many of our customers will be putting out amazing content covering the World Cup, how I will likely spend a lot of time reading this content, and how I’ll be doing my best to ensure that our data will help you deliver your content as effectively as possible. This is when an idea struck me.
Chartbeat should hop on the World Cup bandwagon, and use our data to say something insightful about engagement during the games. Those of you who stare at the Chartbeat Publishing Dashboard all day know that content is consumed differently by visitors from different countries, and as a Chartbeat Data Scientist, I started wondering exactly how engagement with content varied across country. Could I do an analysis to provide some interesting, useful insights into how users from different locations consumed content? And then, I said to myself: Bah! Let’s just have some fun.
So, Chartbeat would like to introduce the Attention Web World Cup, a friendly international competition to see which country has the most engagement with content. We’ll be pitting each country in the World Cup against each other as they consume content on the web during World Cup matches.
How does a match in the AWWC work?
At the same time that a World Cup match is being played, we sample engagement across all of Chartbeat’s sites, filtered by users from the each of the countries that are playing in the current match. About every five minutes during the game, I take the top twenty articles on each of our domains as judged by the number of concurrent users on that article, and then I grab the average engagement time of each of those articles. This is done for each country separately. I calculate a score at the end of 90 minutes by looking at the median of these engagement times. The country with the highest median Engaged Time is the winner.
I thought I’d be fair about this, so I score in a statistical manner, since there is a distribution of times. A draw is possible, if the medians aren’t statistically significant. How did we choose this scoring scheme? Did we have a big meeting here in the Hall of Justice at Chartbeat Studios to decide on the rules of the game? Were there knock-down, drag-out arguments between members of our data science team? No. I just arbitrarily decided this one day last week and stuck with it. So, yeah, there’s that.
So how did the first match of the #AWWC play out?
Match 1: Brazil v. Croatia
And Croatia is the winner! It was quite a close game, but even at the half, Croatia was ahead. Looking at the distribution of Engaged Times shows that engagement was, in fact, quite similar between the two countries. See below: Brazil is in yellow, Croatia is in red.
A larger proportion of Brazilians spent little time engaged on content, and a small percentage of Croatians spent a large amount of time engaged with content. These were the superstars for the Croatian team in the first match of the Attention Web World Cup, and pushed their team to victory.
We can conjecture as to why this is so: Perhaps fans of the Croatian team were scouring sites on the net to find explanations for why those two calls (the penalty and the no-goal) went the way they did, or perhaps Croatians were re-reading all the articles that said Croatia had absolutely no chance of beating Brazil. Were the Brazilian fans more focused on the game, and less on their “second screens?” Whatever the reason, Croatian fans can find some solace in the fact that, at least in the Attention Web World Cup, they came out ahead.
I’m going to be doing this for every match in the Group stage of the World Cup, followed by the tournament style bracket, so keep on checking our blog over the next month for updates on how your favorite team fares. You can tweet me @dpvalente for further discussion about the scoring scheme or anything else related to data/analysis.