Tuesday, July 8, 2014

World Cup: Don't get all defensive!

Saying that the 2014 World Cup is currently underway in Brazil is as obvious as statements come. Even if you're not a huge soccer fan (heck, even if you hate soccer!) it's everywhere, ads, fan tweets, TV results, schedules, preferred links, friends' Facebooks updates with posts as insightful as "Go [insert country name] !![adapt number of exclamation points based on time until next game]".

Given that half of my posts on this blog so far have dealt with NBA basketball you've probably guessed what sport I typically tune in to. But the NBA Finals are over ("Go Spurs!!!!"), and I have to admit I got somewhat caught up in the World Cup excitement.

I have only watched one game from start to finish, but have kept track of results and scores (like everyone on Earth there is a world cup pool going on at work), and something stood out: the first few games were rather exciting (aka high-scoring) (I would sit for short periods of time through games and almost always caught a goal). But the past few games have actually been more on the boring (aka low-scoring) side. Again, associating the number of goals to excitement is quite a shortcut, but I don't know enough of soccer to appreciate the strategic subtleties in a game ending at 0-0 (remember I come from the basketball world were scores average 100). But forget the excitement, is there real a drop in goals scores or is this just a fluke?

Quick soccer recap for the non-experts like me (from Wikipedia):
The current final tournament features 32 national teams competing over a month in the host nation(s). There are two stages: the group stage followed by the knockout stage.
In the group stage, teams compete within eight groups of four teams each. Each group plays a round-robin tournament, in which each team is scheduled for three matches against other teams in the same group. This means that a total of six matches are played within a group.
The top two teams from each group advance to the knockout stage. Points are used to rank the teams within a group. Since 1994, three points have been awarded for a win, one for a draw and none for a loss (before, winners received two points).
The knockout stage is a single-elimination tournament in which teams play each other in one-off matches, with extra time and penalty shootouts used to decide the winner if necessary. It begins with the round of 16 (or the second round) in which the winner of each group plays against the runner-up of another group. This is followed by the quarter-finals, the semi-finals, the third-place match (contested by the losing semi-finalists), and the final.

My naive observation was that scores in the first stage (group stage) were typically higher than in the second stage (knockout stage).
Let's look at the numbers for the 2014 World Cup:

First stage:

  • Average Goals: 2.833
  • First quantile: 1
  • Median: 3
  • 3rd quantile: 4
  • Max: 7

Second stage:

  • Average Goals: 1.917
  • First quantile: 1
  • Median: 2
  • 3rd quantile: 3
  • Max: 3

I won't run any statistical test to look into the difference, but even the worst (best?) devil's advocate will agree that goal production has gone down. The most reasonable explanation is that group stage games don't automatically eliminate you from the competition, everyone is guaranteed three games, so there is a little less pressure to win as opposed to the knockout stage where it is win-or-go-home to use the famous NBA wording. Also, as the competition progresses, the teams get better, it's harder to score 8 goals on your opponent.

But one could have made the case that scores in the knockout stage should have been higher on average: recall that draw (any other sport calls this a tie) are acceptable in the group stage, but a winner is required in the knockout stage and so extra time is added in case of a draw after regular time. The need to break the draw and provide extra time should theoretically generate higher scores.

But is this drop specific to this world cup or has it always been the case? If we break out the scores by stage, does the trend continue with scores getting lower and lower throughout the tournament? Let's look at the evolution of scores for the World Cups from 1990 onwards, broken out by stage:

No strong trends surface from the graph, other World Cups share the same strong decline through the first stages (2002, 2006), but others had increases throughout the same phases of the tournament, as recently as the 2010 World Cup.

Life is full of ironies: as I am writing this post, Germany is leading Brazil 5-0 at the half of the first semi-finals! So much for teams focusing more and more on defense....

So, enough pessimism, the tournament doesn't necessarily generate less goals throughout the tournament, crazy things can happen anytime. Some will say it's the beauty of the sport!

Oh, and based on the graphs, I predict two goals in the Finals....