OR & Sports
What a Kick!
Can an alternative decision-making process
improve World Cup?
By Florence Omosa
The 1998 World Cup is over, but the controversy surrounding the premier sporting event in the world, particularly the format used to determine how teams advance out of the first round, still lingers in the minds of many soccer fans. Judging from the letters written to newspapers in the aftermath of the 1998 Cup, the public is dissatisfied with a system that appears to send some very good teams home after the first round due to a lack of serious play by other teams that have already qualified for the second round. With the 2002 tournament looming, what can FIFA (Fédération Internationale de Football Association), the sports governing body, do to ensure fair play and management of all matches?
This paper proposes a decision-making process that involves all parties with a stake in the tournament. The process utilizes simple models that incorporate both the number of wins, as well as the number of goals scored and conceded by each team. Using the models together with the overall rankings of the teams, the process is designed to overcome the shortcomings of the current system and render match results reasonable to all interested parties.
Like the Olympics, the World Cup is held every four years and attracts a huge audience of international fans. I am one of those soccer fans who gets married to a TV and stays glued to newspapers' sports pages throughout the month-long tournament. During the Cup, my adrenaline level is unpredictable.
However, one thing about the Cup that is predictable is the lackluster performance in late first-round matches of teams that have already qualified to advance to the second round. The situation can adversely affect the chances of other teams to advance. Such was the case in the 1998 Cup when poor performances by Nigeria and Brazil, which had already qualified for the second round with six points, ultimately led to the early sendoff of good teams like Morocco and Spain.
Why is this the case? Why should some teams be in position to consciously influence the outcomes for other teams? Why should several strong teams end up in one pool while several weak teams are grouped in another pool, as is possible under the current system? The phenomenon may lead to a weak team qualifying for the next round while a stronger team goes home. Also, why is a team with six points and two goals considered better than one with four points and six goals? In other words, is the number of wins more important than the number of goals scored? Finally, is it fair for penalties to send a team home?
An Alternative Approach to Decision-making
The 1998 WorldCup tournament will be used to demonstrate the effects of the issues raised above and to exhibit how an alternative process to qualifying may both ensure fair play and satisfy the audiences. Since fans and viewers are central to the success and survival of the World Cup tournament, it is important that they are satisfied with the way the game is played and judged. The decision-making process is critical to fair play and game judgement. It is therefore important that the methods used reflect this.
The proposed decision-making process takes into account the number of wins and goals when identifying qualifying teams. This is achieved through the use of normalized weights for goals and wins. Such an approach is expected to ensure fair play by all teams throughout all the matches. This is because the means to winning will be as important as the win itself in choosing qualifying teams. The specific weight that may finally be used will depend on the interested parties. Sensitivity analysis should be carried out to evaluate the effects of varying weights. The proposal utilizes hypothetical weights for goals (means) and wins (ends).
The alternative method also brings all teams' results together for integrated ranking. This way, the issue of weak teams from a weak pool qualifying for the next round, while strong ones from a strong pool go home, is eradicated. Ties in points can be sorted out by invoking the power of the number of wins, number of goals scored or number of net goals scored (goals scored less those conceded). The proposal employs an additive model that weighs goals scored or net goals plus number of wins. (Those interested in the details of the analysis may contact the author.)
Points (P) = Weighted Goals (G) + Weighted Wins (W)
To reinforce the importance of the proposed alternative process, the method currently in use and variants of the proposed one were applied so as to reveal their effects on the final decision made.
- A. Current:
- (i) Points from wins and draws with a win giving 3 points and a draw 1 point.
- (ii) Integrated - as in A(i) but followed by bringing all the teams' results together and then ranking them.
- B. Alternative Method - it utilizes goals and wins:
- (i) Equal weights for both goals scored and number of wins.
- (ii) Equal weights for both net goals (goals scored less goals conceded) and number of wins.
- (iii) Find corresponding weight (for 0, 1, 2, 3, 4, 5, 6, 7, 8 and 9 goals, corresponding weights were 0, 0.5, 1, 2, 2.5, 3, 3.5, 4, 4.5 and 5, respectively) for goals scored and then add to A(i)'s points (above).
- (iv) Find net goals and then identify their corresponding weight and add to A(i)'s points (above).
- (ii) Equal weights for both net goals (goals scored less goals conceded) and number of wins.
And for 'To Quarterfinals,' only B(i) gives varying results. Adoption of an alternative method at this level will therefore be for boosting perceptions that the matches have been fairly played. Similarly, from 'To Semifinals' onwards, the alternative method does not give rise to varying results. The use of weights here will thus be for enhancing perceptions of fair play and fair decision-making.
On the other hand, when the weights were changed to 70 percent for goals and 30 percent for wins and vice versa, the results remained the same for teams qualifying to the second round of play. Interestingly, the results for quarterfinals and semifinals changed when these new weights were utilized. For quarterfinals, Holland was disqualified and Italy qualified; and for semifinals, Holland qualified and France was disqualified. Strange as it may sound, a 30 percent weight for goals and a 70 percent weight for wins does not change the qualifying teams produced by both the current method and by the one proposed.
Results Altered
All variants of the alternative method, plus the integrated ranking applied to the current method, alter the results of the teams advancing to the second round. A generally dominating feature is the eradication of Chile and Denmark, replaced by Morocco and Spain. The latter teams were heralded by fans and the media as two good teams that undeservedly went home too early.
For qualification to quarterfinals, semifinals and finals, equal weighting and integrated ranking do not affect the final qualifications of the teams. However, the use of such approaches may contribute to fans and viewers perceiving the matches and rankings as having been fair. The methods suggested might also help in reducing the bitterness and helplessness felt when a team is beaten through what is perceived as an unfair play or through penalties. This is because penalties will be weighted like any other goal scored or conceded.
For instance, the win by Argentina against England in the second round or by France against Italy during the quarterfinals is likely to go down well when weights are used. Therefore, though the same teams qualify when the suggested methods are used in quarterfinals through the finals, the overall ranking of teams varies. This is not withstanding the differing results exhibited when the weight for goals is higher than that for wins. This points to the need for thorough sensitivity analysis so that the preset weights to be used are those generally accepted by the stakeholders. The author proposes the use of equal weights for both goals and wins.
I am also in favor of presenting a prize for the team that scores the most goals or records the most wins. Such a prize, combined with the use of the alternative method of determining how teams advance, will encourage every team to play to the best of its ability all the way through the tournament. The new process will also ensure that the better teams, as reflected by goals and wins, will always advance. This means that fans and other viewers will get value for their money. They will also be satisfied as far as the selection of the qualifying teams is concerned. And since satisfied fans and viewers are more likely to support the tournament, the long-term survival of the World Cup will be guaranteed.
| To Second Round | ||||
| Method Teams | New Qualifiers | New Losers | Leading 3 | |
| A(i) | Italy, Norway, Brazil, Chile, France, Paraguay, Nigeria, Denmark, Germany, Mexico, England, Argentina, Holland, Yugoslav, Croatia, Romania | — | — | — |
| A(ii) | As A(i) except for[3] | Morocco & Spain[1] | Paraguay & Chile | France, Argentina, Italy |
| B(i) | As A(i) except for | Morocco & Spain | Denmark & Chile | France, Argentina, Italy |
| B(ii) | As A(i) except for | Spain & Belgium | Denmark & Chile | France, Argentina, Germany & Italy (tie) |
| B(iii) | As A(i) except for | Morocco & Spain | Denmark & Chile | France, Argentina, Italy |
| B(iv) | As A(i) except for | Morocco[2] & Spain | Denmark & Chile | Argentina, Italy & Germany (tie) |
| To Quarterfinals | ||||
| A(i) | Italy, Brazil, Chile, France, Denmark, Germany, Argentina, Holland, Croatia | — | — | — |
| A(ii) | Italy, Brazil, Spain, Morocco, France, Germany, Argentina, Holland, Croatia | None | None | Brazil, Italy, France |
| B(i) | Italy, Brazil, Spain, Morocco, France, Germany, Argentina, Holland, Croatia | None | None | Brazil, Spain, France |
| B(ii) | Italy, Brazil, Spain, Belgium, France, Germany, Argentina, Holland, Croatia | None | None | Italy, Brazil, France |
| B(iii) | Like B(i) | None | None | Italy, Brazil, France |
| B(iv) | Like B(i) | None | None | Italy, Brazil, France |
| To Semifinals | ||||
| A(i) | France, Brazil, Croatia, Holland | — | — | — |
| A(ii) | As A(i) above | |||
| B(i) | Like A(i) | None | None | Brazil, France, Croatia |
| B(ii) | Like A(i) | None | None | Brazil, France, Croatia |
| B(iii) | Like A(i) | None | None | Brazil, France, Croatia |
| B(iv) | Like A(i) | None | None | Brazil, France, Croatia |
| To Finals | ||||
| A(i) | Brazil & France | |||
| A(ii) | Brazil & France | |||
| B(i)-(iv) | Brazil & France | |
||
|
Notes: 1. It is assumed that Morocco and Spain would continue to play as they had before, in other words, better than Denmark and Chile. This means that they would have at least qualified to the extent Denmark did. Otherwise it will be hard to carry on the argument. 2. There is a tie in ranking between Morocco and Denmark. The results of A and B (i), (ii) & (iii) were used to sort it out. Morocco dominated and hence qualified. This may also explain why Denmark is out in A (ii) under "To Quarterfinals." 3. Please note that I have had to come up with hypothetical scores in order to rank under A(i); otherwise it would not have been possible to come up with ranking for it. Also note that the simulation went as far as the new entries are concerned. |
||||
Florence Omosa (florence@pactke.org) is a lecturer in the Department of Management Science, University of Nairobi, Nairobi, Kenya. She is currently studying for a Ph.D. in Decision Sciences at the London School of Economics.
OR/MS Today copyright © 2001 by the Institute for Operations Research and the Management Sciences. All rights reserved.
Lionheart Publishing, Inc.
506 Roswell Street, Suite 220, Marietta, GA 30060, USA
Phone: 770-431-0867 | Fax: 770-432-6969
E-mail: lpi@lionhrtpub.com
URL: http://www.lionhrtpub.com
Web Site © Copyright 2001 by Lionheart Publishing, Inc. All rights reserved.