I just ran a million simulations of the World Cup – Brazil won 450,000 times

Any respectable economist should have a view on who will win the upcoming World Cup in football in Brazil. Therefore I of course also have a view on this.

I have co-authored a paper on this topic with my clever Danske Bank colleagues Jens Pedersen, Morten Thrane  Helt, Stanislava Pravdova and Kristoffer Kjær Lomholt.

This is from the paper – “Brazil set to succeed on the pitch, but it’s an uphill battle for the economy”:

 While there is not much to cheer about regarding the Brazilian economy, we believe the Brazilian population will at least be able to cheer about its national team’s result at the upcoming football World Cup.

In this document, we present our forecasts not only for the Brazilian economy but also for the outcome of the World Cup. We have estimated an econometric model for the World Cup result based on data from the five previous World Cup tournaments and used the model parameters to simulate the upcoming World Cup and the results are clear to us.

In our view, home advantage, a large population and a strong football tradition will ensure that Brazil wins the World Cup. We believe Argentina will be in the running but will lose to Brazil in the final. Germany will take third place.

However, chance is a major factor in football, so nothing is given – not even for Brazil. To describe these factors we have used so-called Monte Carlo simulations to estimate the probability of different teams winning the World Cup. Brazil is strong favourite, with our simulation indicating a 45% chance that Brazil will win the tournament. We calculate the runners up are much less likely to win, with Argentina having an 8.1% chance, Germany 7.6% and France 6.7%.

We believe Spain will be the big disappointment of the tournament and believe it is likely to exit the tournament after losing to Italy at the quarter final stage. We see only a 4% likelihood of Spain winning the 2014 World Cup.


PS I didn’t really do the simulations – my colleagues did.

Follow me on Twitter here.


Leave a comment


  1. Rob Rawlings

     /  May 28, 2014

    For Brazil to win the tournament they have to qualify from a group of 4 (which they seem almost certain to do) and then win 4 elimination games.

    If you give them a 100% chance of qualifying from the group stage then you still have to give them an 80% chance in each of the elimination to get to that 45% win stat. Seems a bit high even for the home (and possibly best) team.

    I’m an England fan though so perhaps I’m just a bit sad that that 3.9% is also an overestimate 😦

  2. even worse, 3.6%

  3. jens

     /  May 28, 2014

    After reading this I am just happy that I’m not a customer at Danske Bank. Bookmakers will surely accept your business based on these estimations.

  4. Bernhard Rohrbacher

     /  May 28, 2014

    If there is only a 45 percent likelihood that Brazil will win the World Cup, then it is more likely than not that Brazil will NOT win the cup. How then can you be so sure that Brazil will win it?

  1. ラルス・クリステンセン「W杯の結果を100万回シミュレートしてみたよー。ブラジルつえー」 — 経済学101
  2. 日本のW杯優勝確率www | プロ野球ニュース
  3. 神速なんJアンテナ » 日本のW杯優勝確率www
  4. An economist take on Brazil-Croatia – completely as expected | The Market Monetarist

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: