Russia’s FIFA world cup 2018 has been one of the most entertaining editions. It has unfolded in the most unexpected ways to have Croatia emerge into the FINALE for the first time in the history of soccer. Fans and supporters (including Airim’s team) of different nationalities have cried their hearts out in joy or lament when they saw the star-studded favourites (Germany, Spain, England) being overpowered by the underdogs (Sweden, Russia, Croatia). Even the greatest of present-day footballers such as Cristiano Ronaldo and Lionel Messi have been unable to carry their teams through the Round of 16.
The most unpredictable of this lot (also most nerve-racking) are the World Cup Finals. The very nature of this unpredictability has made it extremely exciting for the Data Scientists across the world to predict the results based on various models. Like Doctor Strange, they might have already run 14 million simulations to choose 1 winner of the tournament.
Having said that, there have been attempts to predict the winners at each stage of the tournament even when Data science was not around.
Historically, predictions were done by the psychic stars of the animal kingdom. A few examples of which are given below:
2010: Paul the octopus – Spain vs Netherlands (1-0)
Octopus predicting winners by opening its box.
2014: Big Head Turtle – Germany vs Argentina (1-0)
Turtle predicts winners by carrying football to the winning team’s goal post.
2018: The Achilles Cat – France vs Croatia (4-2)
Russian cat eats food from the bowl attached to the winners flag.
It seems fascinating that the above-mentioned oracles have done magically well in their predictions. Although in the present world the prediction accuracy of these these miraculous soothsayers has been beaten to the last percentile using data science algorithms (such as Random Forest).
The take-over of Data Science:
Well…we don’t need to tell you this! Starting from Team line-ups, player attributes, match predictions to making the entire team strategy, data-driven approach has taken over the entire game.
To put this in perspective, in the 2014 edition of world cup, German Football Association (DFB) partnered with software company SAP to develop two new technologies that tapped into the potential of Big Data analytics. As CNBC reports, these two technologies helped Germany to identify strengths and weaknesses of opposing teams ahead of the competition.
1. SAP Challenger Insights:
Provided information on the opposition’s characteristics, their offensive and defensive tendencies and their formations.
2. Penalty Insights Function:
Meant to help goalkeepers and coaches spot patterns around how their opponents take penalty kicks.
While the nations using data science this year are not public yet, we’re sure that France and Croatia will be heating up their analytics engines ahead of the biggest game in the world cup.
Most importantly data science & analytics give a natural edge in football thanks to the highly optimised measuring systems. These systems provide immensely structured data about Possession, Number of Passes, Shots on Target, Running Speed and a slew of other factors. Consequently, the structured data makes it extremely easy to run analytics and get the most accurate results.
This data can be accessed from various platforms such as Kaggle.
But who will win it…?
And finally we’re onto the concluding the winners of the World Cup 2018.
We’ve taken 37 factors into account with every possible permutation. Everything averages out to the probability of France winning by a factor of 62%.
Factors such as current Club, preferred position and even taxation policies of France and Croatia have been taken into account.
We’ve also predicted a score line that will be conceived at the end of extra time.
Who are you supporting in the finals?
Do let us know in the comments!
If you want to learn more about our model of predictions, say “Hello” to our team: