What the Fifa rankings can and cannot tell us about who will win the World Cup

The Fifa rankings reveal something about the mixture of skill and luck that decides a World Cup. A team will probably have to be one of the best to win, but they will probably need the ball to run for them as well.

The compilation of the Fifa rankings changed significantly after the 2006 World Cup, so I studied the three tournaments since then – 2010 in South Africa, 2014 in Brazil and 2018 in Russia.

First I listed the participants in order of their pre-tournament Fifa rankings. Then I listed them in order of their finishing position, which they were given after the tournament by Fifa – one for the winners, two for the runners-up, and so on all the way down to 32 for the team knocked out in the group stage with the worst results.

Then I compared the two sets of numbers. There was a medium-strength correlation between them. Variations in pre-tournament Fifa rankings explained about half of the variations in actual tournament finishing positions.

All three winners and two of the runners-up had been among the top eight in the pre-tournament Fifa rankings. So had three of the six beaten semi-finalists.

The first eight in lists of 32 is the top quarter. So the teams who went furthest were drawn mostly from the teams who were supposed to be most able.

Among qualifiers for Qatar, the top eight in the pre-tournament Fifa rankings are Brazil, Belgium, Argentina, France, England, Spain, Netherlands and Portugal. Those countries fill eight of the top nine places in the betting. The extra country in the betting is Germany. The order of both lists is similar.

There is probably not much disagreement around the world among people who follow football about which countries should be the strongest. Most likely one of them will win. But saying which one, and being right, is harder. Luck can play a big part in a short tournament. No-one plays more than seven games at a World Cup, and half of the teams go home after three.

That is why Louis van Gaal, coach of the Netherlands, one of the eight or nine most plausible winners, said: “I believe we can go far, but we also need luck.”

Position and performance in the EFL

I cannot see any prices I think are wrong for today’s fixtures in League One and League Two, so this seems like a good moment to talk generally about those competitions.

There is a game between Bradford and Northampton who are one place apart in League Two. There is a game at Grimsby who are 12 places below Stevenage. And in League One there is a game at Oxford who are 12 places above Forest Green. Twelve places in a division of the EFL is half the length of the table.

How do the chances of a home win, draw and away win vary with the difference in league position between teams?

I have studied the relationship between results and end-of-season positions. Obviously that is not quite the same thing. Less than halfway through a campaign we do not know precisely where teams will finish. But I will mention a few findings because I think they can provide a useful starting point for thinking about games between teams who seem to belong in different sections of a table.

In games between teams who will finish in adjacent places there are typically 45 per cent home wins, 29 per cent draws and 26 per cent away wins. In games between hosts who will finish half the table above their visitors there are typically 62 per cent home wins, 26 per cent draws and 12 per cent away wins. And in games between hosts who will finish half the table below their visitors there are typically 29 per cent home wins, 29 per cent draws and 42 per cent away wins.

As it happens, those percentages are similar to the ones implied by odds for the games at Bradford, Grimsby and Oxford.