Hardest World Cup groups

[Gwylan]

I thought I'd add up FIFA rankings of all teams in the groups to see which one was the toughest. I know it's a bit rough and ready but it's the only thing that we have - the lower the score, the tougher the group

So we have:
Group D 73
Group C 77
Group H 82
Group B 88
Group E 104
Group A 125
Group G 136
Group F 148

It's interesting to note that England are in the second toughest group and that France got ejected from the third easiest - although Italy could be ejected from the easiest of all.

 

[manintheblackpajamas]

so we have emerged gloriously from the second hardest group to be rewarded with opponents from the hardest group...
it's a fix!!

 

[theonesmith]

Wouldn't it have been better to find the mean score of each group? To stop the scores getting weighed down by extreme scores?

 

[Easy 10]

That is indeed mildly interesting Gwylan.
Good work.

 

[Pantani]

Wouldn't it have been better to find the mean score of each group? To stop the scores getting weighed down by extreme scores?

The mean score would be biased by the same extremes as the total is. The median score may show what you want to know. Although I would argue that the ranking of the third best team probably tells you how hard a group actually is.

 

[Guy Fawkes]

I thought I'd add up FIFA rankings of all teams in the groups to see which one was the toughest. I know it's a bit rough and ready but it's the only thing that we have - the lower the score, the tougher the group

So we have:
Group D 73
Group C 77
Group H 82
Group B 88
Group E 104
Group A 125
Group G 136
Group F 148

It's interesting to note that England are in the second toughest group and that France got ejected from the third easiest - although Italy could be ejected from the easiest of all.

Thats part of the problem with England fans, they ignore rankings, etc... and just go by name, so if we are playing someone like USA they think we should be thrashing them out of sight, but they are ranked 14th and we are ranked 8th - only 6 places higher which is hardly a reason to expect us to thrash them.

We had the second hardest group by World rankings yet the Sun carried the headline after the draw of:
England
Algeria
Slovenia
Yanks
leading to overexpectations of England thrashing them all, anything less and we are deemed to be shit.

 

[Tyrone Biggums]

Well at least we scored 4 points in the hardest group at the WC.

Oh for an easier grouping.

 

[Garry Nelson's Left Foot]

I thought I'd add up FIFA rankings of all teams in the groups to see which one was the toughest. I know it's a bit rough and ready but it's the only thing that we have - the lower the score, the tougher the group

So we have:
Group D 73
Group C 77
Group H 82
Group B 88
Group E 104
Group A 125
Group G 136
Group F 148

I would argue that the ranking of the third best team probably tells you how hard a group actually is.

I thought this was a good idea so I worked it out:
A - Mexico - 17
D - australia - 20
B - Nigeria - 21
H - Switzerland - 24
C - slovenia 25
G - Ivory coast - 27
F- Slovakia - 34
E- denmark - 36

I then thought it'd be interesting to see which group is the closest in ranking between first and fourth. Here's how that turned out:

C- 22 places between 1st and 4th
D- 26
H- 36
B- 40
E- 43
F- 73
A- 74
G - 104

Then, because we've now got three different ways of working out which is the hardest group i thought I'd find out the 'definitive' answer. I gave the highest group in each of the three ranking groups 1 point, 2 points to the next highest and so on. Then by adding up the points scored by each group in the three different ranking groups we can find out which is the toughest.

D- 5
C- 8
H- 10
B- 11
A- 14
E- 18
F and G - 21

So it turns out group we all thought of as the group of death- brazil, portugal and the ivory coast was in fact the joint easiest. As was group that Italy finished bottom of. Meanwhile england were apparently in the second toughest group and will have to play the winners of the toughest group in the next round...

 

[Gwylan]

So it turns out group we all thought of as the group of death- brazil, portugal and the ivory coast was in fact the joint easiest.

I didn't. I thought the group of death was the Germany, Ghana etc - that's how I explained to Mrs G. Turned out I was right but hadn't checked the stats. The other way I could have done it was to take the top-ranked team out and look at the rankings of the three other teams to see how difficult the top's team task was - but this had no effect on the order that I had.

Well done, GNLF for your method. It's interesting to see that, whatever way you cut it, England were in a pretty tough group - despite all the EASY headlines.

 

[Kalimantan Gull]

On a maths point, the first and third methods are virtually the same way of calculating the 'hardness'. i.e. its is completely swayed by how low down the rankings the fourth team is, so you're probably just double weighting this factor.

Also, Group A isn't scoring very high on your methods, because all he teams are quite low down the rankings. But the four teams are all close to each other, making it a very hard group for the teams that are in it.

 

[Kalimantan Gull]

Actually, the FIFA rankings aren't the only rankings out there, there is also something called the 'Soccer Power Index' which I quite like. These are much more fluid rankings than FIFA, so for example in the SPI index France has fallen 15 places to 27th in the last week whereas its still hanging around at 9th in the FIFA rankings.

Anyway, before the World Cup began, the SPI rankings were used to determine the probability of each team qualifying from the group, making the semis and winning the World Cup, as in the table below. Its only maths, but this has England as the MOST LIKELY of ALL 32 teams to qualify from our group. Suggesting that, for England at least, it was the easiest group.

 

[Gwylan]

Actually, the FIFA rankings aren't the only rankings out there, there is also something called the 'Soccer Power Index' which I quite like. These are much more fluid rankings than FIFA, so for example in the SPI index France has fallen 15 places to 27th in the last week whereas its still hanging around at 9th in the FIFA rankings.

Anyway, before the World Cup began, the SPI rankings were used to determine the probability of each team qualifying from the group, making the semis and winning the World Cup, as in the table below. Its only maths, but this has England as the MOST LIKELY of ALL 32 teams to qualify from our group. Suggesting that, for England at least, it was the easiest group.

Interesting stuff JG, I'd never heard of the Soccer Power Index. It's fascinating to see those figures. I can't see how Brazil's had less chance of qualifying than England but there you go. Nor can I see how France's group is seen as one of the hardest. The SPI predicts that Mexico will finish bottom whereas FIFA has them nearly 70 places above SA, there's something wrong there.
EDIT: Just had a look at SPI and see that before the WC started, Mexico were ranked 36 places above SA - really don't see how that translates into SA having more chance of going through than Mexico.

I see that Germany's group is still seen as the hardest though, which ever way you cut it.

 

[Kalimantan Gull]

Interesting stuff JG, I'd never heard of the Soccer Power Index. It's fascinating to see those figures. I can't see how Brazil's had less chance of qualifying than England but there you go. Nor can I see how France's group is seen as one of the hardest. The SPI predicts that Mexico will finish bottom whereas FIFA has them nearly 70 places above SA, there's something wrong there.
EDIT: Just had a look at SPI and see that before the WC started, Mexico were ranked 36 places above SA - really don't see how that translates into SA having more chance of going through than Mexico.

I see that Germany's group is still seen as the hardest though, which ever way you cut it.

On the first point, there is slightly more chance of Brazil finishing third in their group, because of the relative strength of Portugal and Ivory Coast; compared to very little chance (pre-tournament) of England finishing below either Slovenia or Algeria. (although Slovenia's strong showing has led to them jumping 13 places in the SPI in the last month.)

On the second - the table I pasted lists simulated odds based on the Soccer Power Index, together with a few other factors - the strongest of which is Home advantage. So South Africa gets a big home advantage boost, which makes sense because home advantage is VERY important in International football - hence more chance of going through than Mexico - didn't work out that way of course, but it was very close. All other African teams were also given a smaller 'home continent' advantage in these rankings.

If you want to read more, have a look here:

FiveThirtyEight: Politics Done Right: World Cup Simulated Odds
FiveThirtyEight: Politics Done Right: For Information Purposes Only
FiveThirtyEight: Politics Done Right: Is Africa Failing?
FiveThirtyEight: Politics Done Right: World Cup Knockout Round Scenarios
FiveThirtyEight: Politics Done Right: Why We Have U.S.-Ghana at Even Money

 

[seagull_special]

who said maths can't be exciting? great work chaps

 

[rcf0712]

Well at least we scored 4 points in the hardest group at the WC.

Oh for an easier grouping.

and let's not forget that was with one of the best players getting red carded, one after just 25 minutes in each game not won....
My wife and kids are getting me one of those new Socceroo shirts for my birthday next week - should raise some eybrows when I visit England in August, might even wear it to the Withdean for the Rochdale or Walsall game for a laugh....

 

[Gwylan]

On the first point, there is slightly more chance of Brazil finishing third in their group, because of the relative strength of Portugal and Ivory Coast; compared to very little chance (pre-tournament) of England finishing below either Slovenia or Algeria. (although Slovenia's strong showing has led to them jumping 13 places in the SPI in the last month.)

On the second - the table I pasted lists simulated odds based on the Soccer Power Index, together with a few other factors - the strongest of which is Home advantage. So South Africa gets a big home advantage boost, which makes sense because home advantage is VERY important in International football - hence more chance of going through than Mexico - didn't work out that way of course, but it was very close. All other African teams were also given a smaller 'home continent' advantage in these rankings.

Great stuff KG. I think that the home advantage for SA was a bit over-stated - it's important but not that important. The best bit about home advantage is getting seeded but seeing SA ended up with three teams in the FIFA top 17, they got the thick end of the draw there. And I certainly think the 'home continent' advantage is over-stated.

But there's some fascinating stuff there - thanks for pointing it out.

My first job after leaving uni was doing some statistical analysis for a supermarket chain. I did predictions on supermarket turnover based on a variety of factors - how much more interesting would that sort of number-crunching have been if I had analysed football scores?

 

[Pantani]

I have worked out the median value for each group, they are as follows.

A- 16.5
B- 17
C- 19.5
D- 17.5
E- 17.5
F- 32.5
G- 15
H- 21

This method takes out the extremes of the low ranked teams. As a result Brazil's group appears the hardest. Followed by A, B, D and E. This method has England's group as the third easiest, which to me, is about right. Worryingly for Italy they finished last in the easiest group.

 

[Gwylan]

I have worked out the median value for each group, they are as follows.

A- 16.5
B- 17
C- 19.5
D- 17.5
E- 17.5
F- 32.5
G- 15
H- 21

This method takes out the extremes of the low ranked teams. As a result Brazil's group appears the hardest. Followed by A, B, D and E. This method has England's group as the third easiest, which to me, is about right. Worryingly for Italy they finished last in the easiest group.

I don't quite understand the significance of removing the worst team from the equation. It's perfectly possible for the worst team to get a decent result - for example, England beat the second worst team but only drew with the worst. It doesn't strike me as an accurate way of assessing group difficulty.

I think KG's SPI method is best.

 

[house your seagull]

the fifa rankings are skewed aren't they? the concacaf teams always do quite well

 

[Gwylan]

the fifa rankings are skewed aren't they? the concacaf teams always do quite well

Are they though?

The FIFA rankings have Mexico below France and USA below England, yet both those teams outperformed their supposedly superior teams. That doesn't sound skewed to me.

I think the main problem with FIFA rankings is that they don't take into account current results: how can teams like Croatia and Russia who aren't in the Cup be above teams that are - like Slovenia, for example?