AFRICAN CUP OF NATIONS 1998 - After the finals: comments and statistics on the tournament and the performance of the teams

(discourses in) STATISTICS - a part of the African Cup Of Nations Review

• the tables - putting achievement into a little different light
• the regional comparison

• A general review on the tournament
• Review of the performance of the particular teams

If you have a strong team and you would lose only one of six matches, it is also question of luck  whom you have to play against when, especially if it is about play-off matches. If you lose right the first one, you won't have five more to prove you are great. This is why a regular table after points does not display right, how strong teams played in such a tournament. Take for example the unbeaten Ivorian Elephants. Other teams sneak through by luck or with the help of penalty shoot-outs without having really dominated. Whatever you try, you will find, there is no real way to measure the strength of the teams, all ways expose particular aspects and can be questioned for neglecting others. Not to speak of that a performance needs its context, that some teams do easier against particular style of play and worse against a second, while that second again loses to the first. Anyway here are some suggestions to have a different view, some food for your discussions. The view is different from that sense that those tables do not consider for example the four semi-finalists as the four best teams, as you will see. The first idea is always the table after winning percentage. It gives all matches the same evaluation and measures the average of success per game. Traditionally a tie here counts as half a win, when the 3 point rule is more considered as a method to change the attitude than to have a fair way to measure strength.

so here it is:
1) table after winning-percentage (US-notation used, 0.750 = 75 %)
 

	matches	wpct
Egypt	6	0,750
Côte d'Ivoire	4	0,750
South Africa	6	0,667
Tunisia	4	0,625
Morocco	4	0,625
Cameroon	4	0,625
Congo	5	0,600
Burkina Faso	5	0,500
Guinea	3	0,500
Zambia	3	0,500
Togo	3	0,333
Ghana	3	0,333
Angola	3	0,333
Namibia	3	0,167
Mozambique	3	0,000
Algeria	3	0,000

note: third match play off is not taken into account (otherwise: Congo 0.583, Burkina Faso 0.500), matches count without penalty shoot-out
(a table applying the 3-point-rule=counting a tie as a 1/3 win would boost Congo as a no-tie team. Such a table you can see at the result page by Raoul S da Silva Curiel)

You might wonder to find Burkina Faso as good as Zambia. But do not forget, Burkina Faso only won two of its five matches. Anyway this tables leaves you with the idea that advancing to semi-finals should be taken into account as well somehow. This leads to the next solution:

2) table after ®logarithmic winning-percentage
A table for absolute experts. By using the logarithm of the number of games played, instead of the number of games themselves, this table takes into account, that the later matches should have been more difficult than the first ones. It credits the success of the teams that advanced further in the tournament.
 
the first column shows the team, the second the number of matches, the next column then the number of wins devided by the logarithm of the number of matches played to the base of 2.
the last column shows the same kind of calculation but this time to the base of the square root of 2 what is the same as the first value devided by two.
by this trick it is easier to see how this logarithmic calculation affects the table: The teams that advance from group matches to first round but then lose are defined as the average. Their winning percentage does not get affected while the value for teams that do worse gets lowered and the value for teams that advance further gets boosted. You can see this when you compare to the regualr winning percentage table: Morocco, South Africa, Tunisia and Côte d'Ivoire have the same winning-percentage as in the first table, teams that advanced further a higher one, teams that were knocked out in first round, fall back a little.
®© The Shot That Passed Right Through The Net, reprinting of the table or application of the calculation method is allowed but reference to The Shot That Passed Right Through The Net is requested

 

Regional comparison

in the following you will find a short discourse on the attempt to evaluate the achievement of the 5 African regions North, West, Central, East, and South.
You will find that again it is not as easy as it seems at a first look.

The first thing to consider is that the particular regions had different results during the qualification. Due to the fact that those regions contain an uncomparable structure of big, small, rich and poor associations a comparison here can not be based on statistics: the North does not have such a high number of small poor associations so that the percentage of qualified teams has to be higher. What can be said is that there is still a bad situation crisis in the East, when no team qualified for the finals, that there was only one team from Central Africa, DR Congo (in a few cases Cameroon is allocated to Central instead of West) and that the South arrived with a considerable force.
This should be taken into consideration when interpreting the tables, a region that accomplished an overaverage success during the qualification has a higher number of minor teams at the start at the finals and so looks worse in the tables although their overall performance in the competition including the qualification might have been more impressive.

Anyway the following tables consider only the results of the finals.
The first table to offer is the one after winning percentage:
 

	teams	matches	wpct
North	4	17	0,559
West	6	22	0,523
South	5	18	0,389

the East had no team at the finals, the Central only one, what has to be considered as individual rather than representative
 
But note, this table includes the matches north vs north, west vs. west, and south vs. south. Those matches have a winner and a loser and thus shift the values closer to 0,500 mark. Taking only the matches against other regions into account, the differences increase:
 

	teams	matches	wpct
North	4	15	0,567
West	6	14	0,536
South	5	10	0,300

note that the matches against Central (DR Congo) are included here

It becomes obvious that South Africa rescued the region from a disastrous looking result when no other team from the South could win a match against their Western or Northern opponents.

Sometimes teams do easier or worse against a particular style. So the next step is to analyse how successful the particular regions played each other:
 

	matches	wins	wpct
North vs. West	8	3	0,375
North vs. South	6	4,5	0,750
West vs. South	3	2,5	0,833
West vs. North	8	5	0,625
South vs. North	6	1,5	0,250
South vs. West	3	0,5	0,167

remember: tie counts as half a win

You might ask yourself, where the missing matches are and why the West looks so good here but not in the other table above. The answer is easy:
The West lost all 3 matches against Central African DR Congo, while Congo was beaten against Northern Tunisia and Southern South Africa. DR Congo felt remarkably well against the West on Western home turf.
 
Of course this attempt could go far further with tables that take the logarithmic number of matches of the regions divided by the number of teams into account, evaluating the relations to the former position in world ranking to determine whether the performance had been over- or below average, first half vs. second half etc. But more requires a sponsor when this was the best possible to achieve. And I fear Coca Cola did not read it...