Sunday, February 19, 2006

On another note...

Enough about numbers and rankings for a while. Here's a look at the Dutch shirts, home and away, that you'll see at the World Cup. While I don't much care for the collared edition, I would like to have some kind of fan-wear in time for June. Any tips on where to order one of these, cheap, would be more than welcome. Right now I'm leaning towards the away shirt (the white one), but it hardly seems right to be supporting the Oranje wearing red, white and blue. Perhaps there will be some clearance sales on the old shirt, which would serve just as well.

Saturday, February 18, 2006

To be fair...

As much as I like to be critical of FIFA's ranking methodology, and the results that it often yields, I should point out that steps are being taken to improve the situation a little. A FIFA meeting in late 2005 resulted in the following:
Furthermore, the executive body agreed that the FIFA/Coca-Cola World Ranking should be revised. The new system will come into effect after the 2006 FIFA World Cup™ and will include data from the previous four instead of eight years for evaluation purposes.

Dropping the 5-8 year old results is a good first step. However, another article reveals more details (in French):
De points seront attribués pour tous les matchs internationaux des équipes A, suivant qu'il s'agit d'une victoire (30 pts), d'un nul (10) ou d'une défaite (0). On tiendra compte également de l'importance du match (amical, continental, mondial), de la valeur de l'adversaire (son classement Fifa), de son classement régional (ex : Uefa), du nombre de buts marqués ou encaissés, du fait de jouer à domicile ou à l'extérieur.

For those among you who aren't familiar with the Gallic tongue, this paragraph sketches the outline of a system where a team earns 30 points for a win, 10 for a loss and 0 for a draw. Adjustments are to be made based on type of match (friendly, contintental, World Cup), strength of opponent, goals scored and allowed, playing home or away, and the regional coefficient, all of which are more or less part of the picture under the current system.

So, while I like the 4-year window, I still see a few problems with the proposed new ranking. 1. 30-10-0. You see there are no negative numbers. All other factors being relatively equal, a team could play 15 matches in a month, and win 1, and earn nearly as many points as a team that plays a single match and wins. 2. The regional coefficient. If strength of opponent is taken into consideration when calculating the value of result, then the coefficient is entirely useless. If a European match has a higher coefficient (it does), it is because European teams are perceived to be tougher opponents. But the opponent's ranking should itself be an indication of how difficult the match will be.

We'll see how it all plays out. A more detailed explanation of the new system is scheduled during Germany '06, with the first edition of the new (and hopefully improved) rankings to come out in July.

Wednesday, February 15, 2006


The most recent February FIFA rankings provide a lot of insight into how those number-crunchers do business. The most illustrative example is that of the African countries who participated in the just-completed Nations Cup.

Have a look at the participating countries and the places gained or lost:
Egypt +15
Côte d'Ivoire +10
Nigeria +12
Senegal +1
Cameroon +7
Guinea +21
Tunisia +5
DR Congo +5
Angola +3
Ghana +2
Zimbabwe +1
Zambia +1
Morocco -1
Libya 0
Togo -3
South Africa -1

Total places gained: 83
Total places dropped: 5

This kind of data would lead one to believe that a team that consistently played 6 or 7 matches a month, against no better than adequate opposition, at slightly better than .500, would soon be the world's best, according to FIFA (or the statisticians thereof). I know this isn't a zero-sum scenario, but how can there be a net 78-place gain for African sides for playing amongst themselves?

The only change in the top 10 sees Spain dropping into a 6th place tie with Mexico and the U.S. (both were tied at 7th in January).

Thursday, February 09, 2006

African Nations Cup

I won't claim to be an expert on African football (to be honest, I'm not really an expert in anything), but on the eve of the African Nations Cup final, it would seem at least worthy of a blog post.

I'm mostly interested in how the teams performed relative to their world ranking, and also in the results of the five teams headed to Germany '06 (Angola, Côte d'Ivoire, Ghana, Togo and Tunisia).

So, I've crudely ranked the performances of the teams in the ANC, using the following rudimentary formula:
  • top 4 teams are ranked based on the results of the 1st place and 3rd place matches
  • quarterfinal losers take spots 5-8, with group stage points and overall goal difference used as a tie-breaker
  • group stage 3rd and 4th place finishers ranked by points and goal difference
Without further ado, here are the 16 teams ranked, with FIFA ranking given in parentheses (list updated 2/11/2006)
  1. Egypt (32)
  2. Côte d'Ivoire (42)
  3. Nigeria (24)
  4. Senegal (30)
  5. Cameroon (23), 9 pts, +6
  6. Guinea (77), 9 pts, +5
  7. Tunisia (28), 6 pts, +2
  8. DR Congo (78), 4 pts, -3
  9. Angola (63), 4 pts, -1
  10. Ghana (50), 3 pts, -1
  11. Zimbabwe (53), 3 pts, -3
  12. Zambia (58), 3 pts, -3
  13. Morocco (35), 2 pts, -1
  14. Libya (80), 1 pt, -4
  15. Togo (56), 0 pts, -5
  16. South Africa (49), 0 pts, -5
A few quick observations: of the Germany-qualified teams, only Didier Drogba's Côte d'Ivoire has done well which, given the draw, makes me nervous as a Dutch supporter. On the other hand, Togo must be really worried about its chances. It should also be interesting to see how these results affect February's FIFA rankings (due on the 15th). Expect a big jump for Guinea, who beat three higher-ranked sides, and scored a bunch of goals in the process. Côte d'Ivoire should also get a nice boost.

A quick prediction for tomorrow's final: Egypt, with the home crowd behind them, beat Côte d'Ivoire 2-1. update: That was a typo. Of course I meant that Egypt would win in penalties after missing an extra time PK.

Up-late-date: One immediate rankings-related consequence of the ANC is that champions Egypt have moved up to 4th (!) position in the now-even-more-ridiculous mondfoot rankings.