In this article, we look at World Cup predictions based on economic variables and evaluate their success. We also offer some insights on potential alternatives. First we ask whether England’s high dependence on the domestic league for squad selection is a weakness. Second we consider the representation of major European leagues in the latter stages of the tournament.
Before the tournament started, Frontier Economics put out a small bulletin looking at predicting World Cup 2014. This is available on their website http://www.frontier-economics.com/news/against-the-odds-frontier-bulletin-on-the-value-of-teams-at-the-world-cup/
More accurately, Frontier’s analysis looked at the economic value of the major teams and compared this with bookmaker odds of winning the tournament. What we want to do in this blog post is to see how it stood up, now the World Cup is over, and to get some other ideas floating around out there for metrics to explain performance.
The Frontier hypothesis is simple: the more valuable the team, the better players you have and therefore the more likely you are to win the tournament.
A summary of Frontier’s analysis is shown in the chart below.
Source: Football Economics analysis of Frontier Economics data
Frontier draw attention to the fact that the pre-tournament odds from bookmakers did not exactly mirror the economic value of teams. Brazil, installed as pre-tournament favourites, came out 4th in terms of value. Obviously there are a number of things going on here such as home-advantage and distortions caused by one key star player – such as in the case of Messi for Argentina and without him, Argentina would drop to ninth in terms of value.
How accurate was economic value in predicting performance?
Well, clearly there is some predictive power in the economic value of team. 3 out of the 4 semi-finalists were from the top 4 countries in terms of value and indeed the final was played out between the 2nd and 3rd most valuable teams in the tournament.
That said, Spain, England and Portugal (even Italy) are noticeably out of kilter with the predictive power of even a cohort –type analysis ; first-round exits for all of them. This could just be ‘noise’ given that we are only examining one tournament or there could also be something more interesting going on.
Other ideas for explaining performance?
One hypothesis is that it can it can be ‘bad’ to have too many players in a squad playing coming from the respective domestic league. Why? Well, two reasons. Firstly, it means your players tend not to be exposed to the different styles of football played around the globe and secondly, if taken as a signal, it could mean your players are not wanted by other top clubs.
How does this idea fare? The following chart gives the proportion of each major squad that comes from the respective domestic league (i.e. for England this chart shows the % of players in the squad that play in the English league).
Source: Football Economics analysis of FIFA data
Of the nations shown, Germany – the eventual winners – are the only team to progress to the knock-out stages with more than 50% of the squad sourced from their domestic league. Indeed, nations with very high percentages – Russia, England and Italy – all performed poorly. England’s figure of 96%, and Italy’s 87%, are noticeably higher than their German and Spanish counterparts who both have strong domestic leagues. (The very low figures for Brazil and Argentina are more a reflection of the strength of the European leagues compared to their South American equivalents). Of course, this analysis certainly does not suggest that the Premier Leaugue is ‘bad’…
In general, the Premier League was better represented in the quarter-finals of the World Cup than the two main continental competitors with 22% of players coming from Premier League clubs against 9% and 14% for La Liga and the Bundesliga.
Source: Football Economics Analysis
So, summing up. We can reason that:
- Squad value was a reasonable predictor of performance – 3 of the top 4 in the tournament came from the top four most valuable teams. But in this World Cup, economic value was left wanting in explaining some early exits
- The teams with the very high proportions of players from their own domestic leagues fared poorly. Russia, England and Italy all had proportions in excess of 85%.
- The Premier League exhibited the strongest national share, in terms of player representation, in the latter stages of the tournament.
There’s plenty more analysis to be done on explaining National Team performance – for example how under-21 performance and progression matter; whether the number of club games affects performance; how investment affects player quality – and these insights should help national Football Associations enhance that performance.