The penalty of football
You can also listen to the interview [MP3 13.2MB].
The penalty shootouts in football has long been seen as problematic. It decides the game based on the skill and luck of one player rather than the entire team, and is seen by many as necessary but unsatisfying. The outcome it generates is seen as a lottery, rather than a true separation of quality. In 2006, FIFA President Sepp Blatter summed it up by saying "When it comes to the World Cup finals it is passion, and when it goes to extra-time, it is drama. But when it comes to penalty kicks, it is tragedy."
Lets hear the opinions of industry experts. Firstly from Craig Foster. Former Australian footballer, and currently a sports commentator on SBS.
- Craig Foster:
When it comes to the extra-time there is a problem. That's obvious given the fact that FIFA has, over recent years, had to trial a number of initiatives to try and get teams to do something in extra-time and not just wait for the penalties. And trying to incentivize them to attack.
Now from John Didilucia. John is a former football player in the Australian National Soccer League, and is currently the manager of A-league team Melbourne Heart.
- John Didilucia:
In the first instance I think a means to an end is a good way to describe what they are. In the purest form of the game there wouldn't be a role penalty shoot-outs. Because the game will be determined during the 90 minutes. However given a certain percentage of matches in football finishes draws there needs to be a process to determine the winner and who would progress through that different stages of tournaments.
La Trobe University economists Dr. Liam Lenten and Dr. Jan Libich, with their co-author Dr Petr Stehlik from the University of West Bohemia, have been exploring one such alternative. Their hope is that it will alter footballers’ incentives in extra-time, and as a result, resolve the game without the need to rely on the penalty shoot-out.
- Dr. Liam Lenten:
There's some various ideas have been bandied around in recent years is to overcome this problem of a penalty shoot-out and trying to find the winner.
FIFA experimented with what called a "Golden Goal". And that was a case where the first goal in extra-time decides the outcome. That was supposed to mitigate the probability that both teams would score and then you end up going to penalties anyway. The problem with that rule was that it made teams play more defensively than they would have otherwise forget the existence of the sudden-death element or the Golden Goal. Other ideas include taking players off progressively so that by the end of extra-time you only have seven players on the pitch. But that that's really deemed to be unacceptable because you may have a player fatigue being a real factor. One possible idea is to move the penalty shoot-out before extra-time. So you have the penalty shoot-out and then whatever result you get from that, only becomes binding if it turns out that extra-time itself which is played after the shoot-out fails to resolve the deadlock in its own. So in other words we're deciding the tie-breaker before extra-time.
That was Dr Liam Lenten. But what benefit would moving the penalty shoot-out forward have to the game?
- Dr. Liam Lenten:
The idea here is that by moving the shoot-out to before extra-time you are giving players the correct incentive to play attractive attacking football during extra-time. Rather than just two teams feeling each other for 30 minutes hoping to take it to penalties. In a theoretical study by sports economist, Juan Carillo in 2007, he drew this theoretical model that basically said, that having this rule change will have two effects. It would make a team that lost the shoot-out before extra-time and will make this team attack more. On the other hand, for the team that won the shoot-out it will make them defend more. So whether this rule actually leads to better outcomes depends on which of those two effects is dominant. But in the end he concedes that this merely a theoretical model and therefore what we really need to do is to look at the data. And the data really is the only thing that can tell us what will actually happen if you have this rule in place. And that's what we set out to do.
Let’s now turn to Dr. Jan Libich to explain some methodology from their study.
- Dr. Jan Libich:
Basically what we're interested in is the difference between the status quo which is, we have extra time and after that we have penalties. And the difference which would be caused by moving the penalty shoot-out before extra-time. And specifically we want to know how the incentives of the players would change in extra-time in terms of playing more attacking game and in terms of potentially scoring more goals.
So in a nutshell we were interested in comparing two types of situations. One situation is where one team is one goal ahead. And we want to compare that to a situation where the teams are level and there's nothing to separate the teams. And this is precisely the difference that would occur. Currently, at the beginning of extra-time the teams are exactly leveled. There's nothing to separate the teams. Under the new rule, because the penalty shoot-out would proceed extra-time, one team the winner of the penalty shoot-out will have some kind of an advantage which is comparable to one team being one goal ahead. Obviously how do we make it given that this kind of rule has never been implemented? Our main way of doing it is focusing on the first five minutes of extra-time. And basically comparing the outcomes of the game is where there was an early goal in extra-time in the first five minutes. So this is our treatment group. And our Control Group are other games in which there was no goals scored in the first five minutes of extra-time.
And now we're basically looking at the scoring outcomes after that in the subsequent 25 minutes of extra-time. And to compare these outcomes between the treatment group and the control group we basically employ a number of techniques. The simplest technique is basically just looking at the raw data. I can tell you that if we take the treatment group where there was an early goal in extra-time, on the average, after that early goal the game would see a 1.2 goals in the subsequent 25 minutes on extra-time. In contrast to that, the games that are level where there was no goal in the first five minutes of extra-time there will only be an average of about 0.7 goals. So the amount of goals you get is almost doubled if there is an early goal in extra-time.
And we argue that this is kind of precisely what the proposal would do. You would have a penalty shoot-out. Obviously the loser has much stronger incentive to go on and attack. And that is going to increase the probability that the loser scores a goal as well as that the loser concedes a goal. But in either case we're going to see more scoring. You can look at it from a different perspective. And you can look at the proportion of games where there was a goal in extra-time. And you'll see that if there was an early goal within the first five minutes then in 75% of those games you actually have an additional goal in extra-time. Whereas in games where there was no early goal in extra-time you only get about 47% of games when there's actually a goal. So again, this is just a different way of looking at the same thing. An early goal in extra-time increases the probability that you are going to have subsequent goals. And this is because it separates the team and gives one team a stronger incentive to attack which leads to more scoring.
To be confident that these differences are not driven by some other factors, a number of different variables were controlled for in the analysis. Dr Libich explains.
- Dr. Jan Libich:
In addition to looking at the data in this very simple fashion, looking at the descriptive statistics, we actually ran logistic regressions. These kind of regressions have the advantage of being able to control for a number of games' specific factors. Things like home-ground advantage. Things like a momentum. Other features that we control for other types of game. For example, one-legged playoffs and two-legged playoffs. You could argue that this potentially might be different. So we want to control for these difference. We also control for the round of the tournament. So think about the FIFA World Cup potentially if you're talking about the round of 16 or The Finals, that could be different factors of play. Importantly, we also include the bookmaker's odds. So the bookmaker's odds obviously already incorporate a lot of the games' specific factors about one team being the favorite. And home ground advantage and so on. So by including that, that also controls all these additional factors. So having done that, that gives us high confidence that the results we're finding are not due to any of those factors that we would have omitted. But they are really due to the rule change that we're trying to study.
With games analysed, data crunched and variables accounted for, the results give a clear indication on what staging the penalty shootout before extra time could bring.
- Dr. Liam Lenten:
The results tell us that by moving the shoot-out to before extra-time, would end up with more attacking, attractive football during extra-time. One way to look at it is how many goals are likely to be scored in extra-time both with this rule changed and with that this rule changed. The results indicate that we will end up with more goals on the average if this rule would have been brought in. We get a similar story if instead we look at the problem as being the probability that the match will be decided without penalties becoming binding. And we also get the same result if instead we look at it as the probability that at least one goal will be scored in extra-time. It doesn't matter which way you look at it. The results are unanimous.
- Dr. Jan Libich:
I was intrigued at the proposal when I first heard about it. But as a researcher we were all very cautious. These kind of rule changes would have quite a lot of impact on the players and a lot of people. So you really want to be sure that they would have the desire-type of incentives. For example the odds of a goal being scored in extra-time would triple under the new rule putting it in terms of the probabilities. Currently, the probability that there will be a goal in extra-time, again, it depends on the various circumstances. For example, look at the FIFA World Cup. And think about a game of a quarter final stage where it was 0-0 in regulation time. The probability that there will be at least one goal in extra-time is about 35%. Under the new proposal the probability would go up to about 62%. So this is a 76% increase in the probabilities.
Change never comes quickly, nor is it always welcome, but in this study it is shown there would be a definite improvement to a game of football. It would be exciting, and a fairer test of skill. Let’s hear from our four contributors, and see what they think the implications of such a change would be.
- Dr. Liam Lenten:
This result could have made to a different results in the future matches than what we will see under the current rules. But the desirable outcome is that there's a greater chance of the result being decided in an open-play or a normal play rather than by penalty shoot-outs. And that's what we're looking for.
- Craig Foster:
If you are going to change anything you change it as little as possible. And all you are doing is changing the dynamics not the rules per se. And so all you do is move the penalty shoot-out. You wouldn't change it.
- John Didulica:
So I think one of the downsides it brings is it disjoints the game. So you would have a 25-minute period where players get cold. Get out of the game rhythm and you have to play again. And I think that is an issue that would need to be managed. So physically, mentally they're not in in tune with the game anymore some of them might be rusty. But I think where it is beneficial is it then adds an extra technical layer to the match. So whilst there's a penalty shoot-out it's not entirely determitive of the outcome. There is a chance the team that looses to redeem themselves in the game of football. And that adds a massive technical layer to the… in the lasagna that this football.
So you might lose in the shoot-out and then you might re-reconfigure the way you're playing to score a goal. And likewise attain whose won the shoot-out might then adopt a different technical approach to getting more plays behind the ball and holding their lead. Well the thing that's worth examining, I think, is you need to find out is how will people react to different incentives. So I think it's certainly worth assessing and worth experimenting with.
- Dr. Jan Libich:
The main change would be that… that would probable be less randomness in terms of the winner. The stronger team would probably be more likely to win. The reason being that again the game are largely going to be decided in open-play. There's going to be less relying on penalty shoot-out. Because of the perverse incentives that have created by the current rule, players play too cautiously in about 50% of all extra-times as scoreless. So there's not a single goal in extra-time. And obviously all those games are decided in a penalty shoot-out which is largely at random. The outcome doesn't really reflect the quality of the teams. Now, under the new rules, we estimate that the proportion of games where there's no scoring in extra-time would be reduced from 50% to less than 25%. So that would pretty much half the amount of games where the penalty shootout actually decides.
So let’s offer up a quick prayer to the Holy Zebra that Australia might win the FIFA world cup, and say thanks to our guests today, John Didulicia, Craig Foster, Dr. Jan Libich and Dr. Liam Lenten. We’ll leave the final words to legendary English football manager Bill Shankly. "Some people believe football is a matter of life and death. I'm very disappointed with that attitude. I can assure you it is much, much more important than that."