by Jeff Dieffenbach

Since August of 1993, FIFA has been drawing up monthly rankings of the world’s national teams. On the eve of the December 1 draw in Busan, Korea Republic to determine the groupings for the upcoming 2002 playing of the 17th World Cup, the latest round of such rankings are now complete.

The rankings consider a broad range of matches over the previous eight years: World Cup qualifiers and finals, continental championship qualifiers and finals, Confederations Cup matches, and international friendlies. For these matches, rankings factor in game outcomes (both result and score), location, and importance, plus opponent and regional strength.

On the surface, a reasonable approach, although some might argue that the eight-year window is too long. While only ceremonial in nature, it is worth holding these rankings up to the light of actual World Cup results.

In 1993, Germany, Italy, Brazil, and Norway topped FIFA’s December rankings. All four qualified for the 24-team 1994 World Cup in the United States, but number six Denmark did not. Other teams ranked in the top 24 that failed to make the finals included France (15), Uruguay (17), Czechoslovakia (19), Portugal, Colombia, and Scotland. Note that all are European or South American.

For the teams that did qualify, did their FIFA ranking correlate to World Cup performance? A decided no. In fact, the correlation between seeding based on FIFA ranking coming into the draw and final team performance was a pitiful 0.35 (For details of how this analysis was conducted, see the discussion at the end of this article. Readers interested in learning more about statistics might consider getting out of the house more.)

Who rose above expectation, and who failed to perform? Powers Brazil and Italy played true to form, and well: Brazil came in seeded 3rd and proceeded to win the tournament (+2 spots); Italy arrived 2nd and finished 3rd (-1).

Bulgaria and Saudi Arabia emerged the proudest of their performances, rising from 20th to 4th (+16) and 22nd to 9th (+13), respectively. On the disappointing end, Norway fell from 4th to 13th (-9) and Colombia plummeted from 9th to 18th (-9). (Recall that Colombia’s woes would soon turn to tragedy with the subsequent gambling-related murder of star Andres Escobar.)

In 1997, FIFA’s December favorites included Brazil, Germany, the Czech Republic, and England. Missing again from the 32-team tournament in France was the Czech Republic (FIFA rank of 3), joined on the sidelines by Russia (12), Sweden (18), Bolivia (24), Ecuador (28), Portugal (30), and Egypt (32). Again, Europe and South America took the bulk of the absences.

Measured against actual performance, FIFA’s rankings predicted actual results even more poorly than in 1993 with a dismal correlation of 0.30.

As in 1993, Brazil and Italy once more played true to form, and once more played well: Italy entered as the 8th seed and finished 6th (+2); Brazil came in 1st and finished 2nd (-1). Likewise, France triumphed from the number 5 spot (+4).

Nigeria outplayed its seeding by the largest margin, moving from 32nd to 11th (+21). Croatia, though, was the surprise of the tournament, climbing from the midpoint (16th) all the way up to number 2 (+14). Other surprises? USA’s fall from 21st to a tie for last (31, -10) and Japan’s collapse from 12th to 31 (-19).

What of 2001’s rankings? The Czech Republic misses out for a 3rd consecutive time (rank: 13), along with Colombia (5), The Netherlands (8), Yugoslavia (12), Romania (15), Norway (25), Honduras (26), Iran (29), and Trinidad & Tobago (tie 31). Once again, note the hit suffered by Europe and South America.

The last entry in the “missing” list highlights the failings of the FIFA rankings. How can Costa Rica, a strong winner of CONCACAF qualifying with 23 points in 10 games, be tied at 31st in the world rankings with Trinidad & Tobago, who finished last in the same qualifying group with only 5 points? The answer is quite simple—the (inflated?) importance of past results. Only a year ago, Costa Rica’s world ranking stood at 60 compared to T&T’s 29.

Brazil (seeded 3rd) and Italy (6th) would seem poised to end among the tournament’s leaders. France also, as the top seeds have fared well (Brazil finishing 2nd in 1998 and Germany finishing 5th in 1994).

Only the games will determine the rising stocks and falling stars. At number 20, the United States has room to move … in either direction.

Next: Why do Ranked Teams Fail to Qualify?

# # # #

How the analysis was conducted: The author examined FIFA rankings for December of 1993, December of 1997, and November of 2001 (the end of the year prior to the following year’s World Cup final). The qualifying field was then seeded based on these rankings.

The difference between seed and rank reflects teams that did not qualify. For instance, if the third ranked team did not qualify (as was the case with the Czech Republic in 1998) but the first, second, and fourth did, the 4th ranked team would have its World Cup seed elevated by one to 3rd due to the absence of the third ranked team.

Final World Cup performance was based on accumulated points: three for a win, one for a draw, none for a loss. No accounting was made for goal differential or match importance. For instance, Brazil and Croatia tied for second in the 1998 World Cup based on 2-1-0 record in their groups and three subsequent wins for a total of 15 points. Brazil’s semifinal win over The Netherlands and its loss in the final to Brazil is given equal weight to Croatia’s semifinal loss to France and its bronze medal (3rd place) win over The Netherlands.

Correlation was arrived at by performing a linear regression of seeding versus actual performance.