Improving Football with Analytics

I have a deep passion for football – or, soccer as it is called in the United States. On Friday, the FIFA World Cup, watched by billions around the world, kicks off with a game between the team of the host country – South Africa – and Mexico. So in this post I want to meld my business passion – analytics – with my sports passion.

Many people are using models to figure out the potential World Cup winners; from PWC’s recent study to Sorbonne professor Wladimir Andreff ,whose model gives Germany a 96 percent chance of making it to the semi-finals (for my own predictions, click here).
 
But no matter how much noise is made around the event, many Americans still take little notice.  Soccer’s seemingly slow pace deters some, others say they don’t understand its rules (if you are the latter, here is a guide with all 17 rules listed, including a tutorial for rule 11, the offside rule).
 
Soccer is very different from American football or baseball.  The plays in American football and baseball are determined and measured in clearly defined increments or “sets.” Soccer, by contrast, is more fluid, with few interruptions. American football and baseball fans can easily understand the impact of a team’s tactics because they are broken up and highlighted play by play. By comparison, some new soccer spectators become bored. They begin to see soccer as a series of uncoordinated activities that sometimes succeed but usually fail.
 
What can be done to get more people to understand the game?  The answer, I think, lies in “analytics.”  Spectators need the right ones, and more of them. 
 
Everyone knows that the way to keep score in soccer is the number of goals.  Goals, however, are an imperfect measure of overall performance, amplified by the fact that not many are made in the course of a game.  Should the performance of a team be appreciated by the final tally only?  Or perhaps, is there more to it?  In the business world, has anyone ever tried to run a company with minimal visibility until the numbers are in? 
 
What’s interesting is that the measures necessary to appreciate a game of soccer and players’ performance are known and available.  Coaches and experts use them, but they are not well communicated to the general public.  Just like in business, until the factors of performance are well-communicated, true understanding of performance is left to a few.

To correct that, broadcasters should publish more analytics on the screen. Commentators also should spend more time discussing the analytics and use their experience to help viewers discern correlation, causality, and coincidence between activities and results.  The typical “ball possession” and “shots on goal” metrics are often mentioned. But how about smarter analytics such passing accuracy, consecutive touches, interception averages, just to cite a few examples?
 
Here are two analytics to consider:

  • “Who’s the boss” analytics: The player who touches the ball the most amount of times and/or provides the largest number of assists for shots on goal.
  • “Penalty” analytics (or should I say “penalytics”):  Probability of a player to succeed on a penalty kick based on his and the goalie’s past performances.   According to ESPN Insider, players have the highest probability of scoring by shooting to the top left or right section of a goal.   Can we get this information for each player before they take a shot?

Each of these metrics could be monitored in real-time and compared to the team’s and/or comparable teams’ averages.  Further, commentators should use them during the game to make predictions, such as predicting at the hour mark who will win the game by the end of the period.  Take a look at this interactive infographic from Chiqui Esteban, Director at lainformacion.com.  It tracks ball movement in last week’s game between Spain and South Korea.  You’ll notice the large number of passes going to player #22, Navas.  Could this have been a leading indicator of Navas’ winning goal?
 
A lot of the data is available and it should move beyond the closed circle of experts (see what leading clubs like A.C. Milan do with their data here).  When data is better shared, when the right analytics are communicated and more often, the game becomes more interesting to more, and better understood by most. 
 
You can find more videos and other tools that will help you experience a more “analytical” World Cup @ www.microsoft.combi.  Also make sure to visit www.extendedresults.com/worldcup, a Microsoft partner that is building social media analytics for the World Cup. You’ll see top mentioned players and many other stats.
 
I look forward to your feedback on this post or my world cup prediction @ [email protected].  Until then, let the games begin and may the best team win…well, you know what I mean…

Posted by Bruno Aziza
WORLDWIDE STRATEGY LEAD, BUSINESS INTELLIGENCE