World Cup Predictions: The final countdown

By: Patrick W. Zimmerman

Just 4 teams left!  Only the cream of the crop ever gets to this stage of the World Cup.  One expects to see Germany, Italy, Brazil, the Netherlands, Argentina some years.  You know, the biggies.   In 2014, there were 10 titles and 11 2nd place finishes between the 4 semifinalists.  In 2010, 5 and 7.  In 2006, 7 and 6.  You get the idea.

Wait, it’s France v. Belgium and England v. Croatia, with a combined 2 titles and 1 runner-up finish between them?  What insanity is going on, here? Not even Putin was expecting Russia over Spain, sending PM Dmitry Medvedev instead of attending the match.


The question

So, given the insanity of certain aspects of the knockout rounds, how did our predictions do compared to the competition?

Not bad, actually!

….and who’s going to win?


The short-short version

While our overall score in the tournament trails a bit behind the FiveThirtyEight and the Odds line, our model got 3 of the 4 semifinalists right, mostly because it ever-so-slightly rated Belgium over Brazil.

Yeah, we’re going to focus on that and ignore the fact that we picked Spain as finalists on the other side because so did everyone else.

Let’s see what the model says about our final four. ::computing:: ….. model says: France over Belgium and England over Croatia in the semis. France over England in the final and Belgium over Croatia in the 3rd place game.

Now we just need France to take it home, and we’ll be golden.


Model performance in the knockout round so far

Knockout qualifier picks
Model Games correct Should have picked Instead of Semifinalists Champ alive?
Principally Uncertain 8/12 Russia, Sweden, Uruguay, Croatia (QF) Spain (2x), Switzerland, Portugal 3/4 Yes
The Betting Odds 8/12 Russia, Sweden, Belgium (QF), Croatia (QF) Spain (2x), Switzerland, Brazil (QF) 2/4 No
The FiveThirtyEight 7/12 Russia, Sweden, Uruguay, Croatia (QF), Belgium (QF) Spain (2x), Switzerland, Portugal, Brazil (QF) 2/4 No
The Power Rank 6/12 Russia, England (2x), Uruguay, Croatia (QF), Belgium (QF) Spain (2x), Colombia (2x), Portugal, Brazil (QF) 1/4 No
FIFA 4/12 Russia, Sweden, Uruguay, Croatia (2x), France (2x) Spain (2x), Switzerland (2x), Argentina (2x), Portugal, Brazil (QF), Denmark 0/4 No

Looking pretty good here! We’re the only group with our champions still alive (France), and that Belgium v. Brazil game really helped us out, since every other ranking system went with Brazil to win.

Spare a chortle for FIFA’s optimism regarding Argentina and Switzerland, though their rankings obviously did not take into account James Rodríguez’s injury convincing Colombia decide that a hack-a-thon (with 34 fouls between the teams!) was the best strategy against England.


Model performance since the beginning of the tournament

Model scoreboard
Model Points Points % Correct results Correct % Ties predicted Tie %
Betting Markets 37 0.638 33 0.569 0 0.00%
The FiveThirtyEight 37 0.638 32 0.552 4 6.56%
Principally Uncertain 36 0.621 30 0.517 8 13.11%
FIFA rankings 33 0.569 27 0.466 15 24.59%

The spread between the models have tightened as the tournament has gone on, which makes a lot of sense given the reduced number of teams and the type of upsets that have happened during this tournament.

There have been few mild upsets (Sweden over Switzerland, Uruguay over Portugal, Belgium over Brazil), which models tend to disagree over (and therefore create some differential between them), and one side of the bracket was pretty much thrown into total chaos by Spain going down in the first round (the universal pick to make the finals from that side), which everyone got wrong.

In short, this was a tournament where there was a pretty tight consensus about who was good (Brazil! Germany! Spain! Argentina!), and in which that consensus was just wrong. That tends to create a clustered group of models.


What next?

3 more games left!  Enjoy them while we can, then tweak the model for the Women’s World Cup and the Copa América next year.

About The Author

Architeuthis Rex, a man of (little) wealth and (questionable) taste. Historian and anthropologist interested in identity, regionalism / nationalism, mass culture, and the social and political contexts in which they exist. Earned Ph.D. in social and cultural History with a concentration in anthropology from Carnegie Mellon University and then (mostly) fled academia to write things that more than 10 other people will actually read. Driven to pursue a doctorate to try and answer the question, "Why do they all hate each other?" — still working on it. Plays beer-league hockey, softball, and soccer. Professional toddler wrangler. Likes dogs, good booze, food, and horribly awesome kung-fu movies.

No Comments on "World Cup Predictions: The final countdown"

Leave a Comment

Your email address will not be published. Required fields are marked *