Historical ratings of international football teams indicate that this weekend will see the most evenly matched World Cup final ever. The game could well go into extra time – and if it ends up with penalties, evidence suggests that the captain who wins the toss gives his team the edge.
On Sunday, France and Argentina will contest the 22nd World Cup final in Qatar. It’s been a long journey to this point, with two years of qualification matches, many affected by the Covid-19 pandemic, culminating in the one-month finals tournament in which 32 teams have been reduced to just two.
Those two, Argentina and France, were the third and fourth ranked teams at the outset of the tournament, behind Brazil and Spain. Both have, as a result, closed their gaps on the top two, who exited the tournament in the quarter-finals and Round of 16 respectively.
Furthermore, because France have been marginally more impressive in the tournament (needing to beat England, the fifth ranked country), they have closed the gap on Argentina such that there is only three points between them in the rankings – in other words, the final is about as evenly matched as is possible.
The Elo prediction is 50.4 in favour of France – not quite 50/50, but close enough at 50.4/49.6. This can be interpreted as a 50.4% chance of France winning the final, and a 49.6% chance of an Argentine triumph.
Figure 1: Elo rating
Source: Author's calculations
This makes Sunday’s decider the most evenly matched World Cup final in history, just beating Italy’s match-up with Czechoslovakia in 1934 (51.9). The pairing of 1934 with 2022 may seem somewhat strange, but the awarding of World Cups to Mussolini’s Fascist Italy in 1934 and Qatar in 2022 do bear some striking similarities.
Because Elo ratings are calculated based on match results alone, they can be calculated for as far back as there has been international football matches. Of the teams in the 1934 World Cup, Austria (semi-finalists) was the strongest, followed by Argentina (who lost in the last 16), and then Italy, Spain (controversially beaten in a replayed quarter-final) and Czechoslovakia. Hence, a lot like this year’s edition, the strongest teams did not necessarily make it through an intense knock-out competition, and yet despite that, there was a very equally matched final.
The most recent close final was 2014 between Germany and Argentina (53), which the Germans edged 1-0 in extra time. France’s final in 2018 against Croatia (61.9) isn’t the most unbalanced World Cup Final either. Germany’s 1-0 win over Argentina in 1990 (65.5), Brazil’s 3-1 win over Czechoslovakia in 1962 (67.4) and Germany’s 2-1 win over the Netherlands in 1974 (69.7) are the most unbalanced finals in history. Arguably, in these three cases, it is the supremacy of Germany and Brazil at those points in time that led to the finals being relatively unbalanced.
Will this weekend’s game be decided in 90 minutes? Thirteen finals have been decided in regulation time, but seven have gone into extra time. Five of those were among the closest historically: Italy-Czechoslovakia in 1934; Germany-Argentina in 2014; Brazil-Italy in 1994; Argentina-Netherlands in 1978; and England-Germany in 1966. None of the ten finals with the biggest imbalances went to extra time, let alone penalties.
All of which is to suggest that we may get extra time on Sunday. Penalties are less frequent in World Cup finals. They were first introduced into the tournament in 1970, but in the 14 competitions since then, just two finals have been decided by penalties – Brazil’s triumph over Italy in 1994; and Italy’s win against France in 2006. In that time, five finals have gone to extra time.
Researchers have long tried to make sense of penalty shoot-outs. While there is a widespread perception that they are a ‘lottery’, that view is somewhat challenged by persistent patterns, such as England’s struggles prior to 2018, and Germany’s pre-eminence in the format. But these are small samples, and there have been many shoot-outs.
The basic observation is that the team taking the first penalty has an advantage, as much as 60/40, in a shoot-out. But a more recent study has shown that this is actually related to the decision a team’s captain makes. The captain that wins the toss before a shoot-out chooses whether to go first or second, and could decide to shoot second (say if they have a strong goalkeeper like Argentina’s Emi Martinez). Irrespective of whether a team shoots first or second, it is the team that wins the toss and hence gets to decide that has a 60/40 advantage in penalty shoot-outs.
Sunday’s final is likely to be close. Indeed, it’s the closest final ever, and most close finals are tight affairs, many going to extra time. But if we are in extra time, there’s still only a 40% chance we get to a penalty shoot-out based on historical patterns. And if we get there, look out for which team’s captain wins the toss – that could be just the edge they need to get their hands on the World Cup.
Where can I find out more?
- Evaluating strange forecasts: The curious case of football match scorelines – study by James Reade and colleagues.
- Going with your gut: The (in)accuracy of forecast revisions in a football score. prediction game – another research paper by James Reade and colleagues.
- Futbolmetrix’s blog: Discussion of football (futbol, calcio, soccer) and numbers.
- Psychological pressure and the right to determine the moves in dynamic tournaments – evidence from a natural field experiment: Study by Dominik Schreyer and colleagues, which finds that teams whose captains can decide about the shooting sequence for penalties are more likely to win the shoot-out.
- Psychological Pressure in Competitive Environments: Evidence from a Randomized Natural Experiment: Study of penalty shoot-outs by Jose Apesteguia and Ignacio Palacios-Huerta.
- Beautiful Game Theory: How Soccer Can Help Economics: Book by Ignacio Palacios-Huerta, which includes his research on the economic analysis of penalty kicks.
Who are experts on this question?
- Alex Krumer, Molde University College
- James Reade, University of Reading
- Daniele Paserman, Boston University
- Dominik Schreyer, Otto Beisheim School of Management
- Ignacio Palacios-Huerta, London School of Economics