TEAMS from 15 countries are arriving in England ready for the start of the European Football Championships next week. For team coaches like England鈥檚 Terry Venables the last few frantic weeks of training have been vital. He will have ensured that his players are fit, that they play in the right formation and that, as they say in football-speak, 鈥渢hey know where the back of the net is鈥�. It is the textbook coaching that any manager would employ before an important tournament. But Venables might be surprised to learn that there is another tactic open to him. He could invoke chaos theory.
Put simply, chaos theory shows that apparently random phenomena have an underlying order. It can describe the behaviour of all manner of systems, from a dripping tap to weather patterns and the creation of galaxies, stars and planets from the maelstrom of the big bang.
Now sports scientist Keith Lyons and his colleagues at the University of Wales Institute in Cardiff are exploring the notion of football as a chaotic system. Lyons鈥檚 team argues that although a football match may appear to be a random affair on one level with 22 players able to run and kick the ball anywhere on the pitch, it is actually highly structured, with both teams following well-established patterns of play. The researchers treat football as though it were a chaotic system. This allows them to understand the game and the patterns it evolves from a fresh perspective, according to team member Mike Hughes.
Advertisement
The Cardiff researchers insist that applying their brand of chaos theory is not necessarily a prescription for success鈥攊ndeed considerable controversy remains over its uses. But they argue that it can give a new angle to coaches and teams trying to understand how best to turn a match to their advantage. 鈥淪uccessful teams exhibit patterns of play,鈥� says Hughes. 鈥淭heir actions, such as passing, defending and shooting, show invariance which provide football matches with a pre-ordained course. Yet amid the patterns there are perhaps four or five occasions when the game deviates from this rigid structure.鈥� Hughes and Lyons believe that if teams can create and exploit those variants or 鈥減erturbations鈥� they are more likely to win. 鈥淥ur evidence suggests that winning teams actively intervene in games to maximise their chances and only the best teams can break the game鈥檚 structural restraints,鈥� says Hughes.
And this is only one application of chaos theory in football. Off the pitch, the Cardiff researchers have used it over the course of a tournament or through one or more seasons to study the evolution of teams and their long-term strategies, successes and failures.
Lyons and Hughes say that on the pitch, the key to cashing in on chaos in the short term is to understand the patterns that exist within a match. For more than a decade, Hughes has been collecting data from matches using notational analysis, a technique for transferring information about a game from video onto computer-generated grids. He notes every action taken by the players鈥攚here they run, pass, how far they kick and so on. From this, the researchers model a team鈥檚 patterns of play.
From data gathered during the 1994 World Cup, for example, it became clear that the Dutch team were aiming many cross-field passes from the left side of the field towards the opposition鈥檚 far goal post in an attempt to set up scoring chances. Similarly, Lyons and Hughes have data that show the styles and movements of individual star players. Bulgaria鈥檚 Yordan Lechkov and Romania鈥檚 Gheorghe Hagi were dominant during the 1994 World Cup and notational analysis shows that they have distinct patterns in their movements, passes and shots.
Lechkov, for example, often picks up the ball on the edge of his own penalty area and runs in an arc on the right side of the field, heading for the opposition鈥檚 penalty area. 鈥淭hese are the patterns these players feel comfortable in and with which they create dangerous attacks,鈥� says Hughes. 鈥淲e can use this information to enable opponents to upset these patterns.鈥�
Those moments in football, where the defence is left gawping in amazement by an unexpected move, are legendary. One celebrated example took place in 1970 at a free kick taken by Coventry City鈥檚 Willie Carr, who stood with the ball between his legs and flicked it up behind him with both heels. His team-mate Ernie Hunt volleyed it straight into the goal, 20 metres away. 鈥淭hat was a classic perturbation,鈥� says Hughes. 鈥淭he unexpected action that is impossible to defend against, mobilised to a team鈥檚 advantage.鈥�
A more recent example took place in the final of last year鈥檚 European Cup Winners鈥� Cup. In the final minute of extra time, Real Zaragoza鈥檚 midfield player Nayim noticed that Arsenal鈥檚 goalkeeper, David Seaman, was slightly out of position. In a completely unexpected move, he took a shot from the halfway line, 50 metres out. The ball sailed into the net to win Zaragoza the cup.
There have been plenty of other crowning moments, when the dribbling of George Best or an accurate pass from Pel茅 has cut through a defence like a knife through butter. But where鈥檚 the need for chaos theory? Surely coaches and even armchair supporters would have to be blind to miss these events?
鈥淭hat鈥檚 true,鈥� admits Lyons, 鈥渁nd by the same token most would want Hagi, Pel茅 or George Best in their team.鈥� The big advantage of notational analysis, he claims, is that it quickly identifies what is obvious or invariant in a football match. 鈥淲hen you can see the orderliness behind football and winning, it is easier to identify the ripples of player-induced chaos that upset these patterns,鈥� says Hughes.
There鈥檚 another benefit too. Those key moments in football that every fan remembers are only the most obvious perturbations. Sometimes critical incidents are much less obvious鈥攑erhaps a short run by an attacker taking a defender 5 metres closer to the wing, allowing space for a second attacker to run inside. These things only emerge in post-match analysis.
鈥淎 coach might instruct his defenders to watch Hagi so closely that they free up his skilful Romanian team-mate Dan Petrescu,鈥� says Hughes. 鈥淥ur data would have identified Petrescu鈥檚 patterns too.鈥� That way, a coach would not fall into the trap of concentrating too much on one player.
Explaining the weaknesses
Once you鈥檝e spotted a pattern in the opposition鈥檚 play, the next step is to try to exploit it. That鈥檚 the task that the Cardiff researchers have now embarked on. 鈥淚t鈥檚 a simplistic example, but the data from the Dutch matches show that the goalkeeper鈥檚 kicks mostly land within a closely defined area of the field,鈥� says Hughes (see diagram). 鈥淭eams defending such kicks could maximise possession by placing a defender where the kicks are landing, so creating a minor perturbation.鈥�
Of course the best way to unsettle the opposition is to buy a great player such as Hagi, who creates his own perturbations. But most teams can鈥檛 afford such a player. Instead, Hughes has great hopes for the techniques he is working on. 鈥淧erhaps we can identify how teams might move closer to playing like Hagi if they can understand the perturbations he causes and how they affect the outcome of a game.鈥�
The researchers are keen to stress they are not trying to tell coaches what to do. 鈥淪ports science has often been arrogant in presuming it can tell coaches to do things that are irrelevant to their circumstances,鈥� says Lyons. 鈥淏ut our studies allow us to see vital cues that may be missed and this, allied to the tacit knowledge of an expert like Venables, could prepare us for a game of football that nobody has yet experienced.鈥�
Jean-Francis Grehaigne of the University of Franche-Comt茅 in Besan莽on, France, studies the evolution of configurations and structures in football teams and matches. He is excited by the theoretical work being carried out in Cardiff. 鈥淭here are many dynamically interacting elements that make up the varied patterns in a football match,鈥� he says. 鈥淎ttempts to break the balance between two teams can provide the temporary imbalance favourable to the scoring of a goal.鈥�
A coach who is prepared to risk introducing a variable into his team鈥檚 style of play could upset the opposition鈥檚 tactics while mobilising aspects of his own team鈥檚 structure. So, for example, switching a player from defence to attack halfway through a game could upset the opposition鈥檚 pattern of play and free another attacker to create more goal-scoring chances.FIG-20333801.jpg
The notion of chaos in football is not limited just to the 90 minutes of the game, however. Hughes and Lyons have also studied the way perturbations affect teams over the course of a tournament, season or longer.
Most teams find periods of equilibrium鈥攕ome successful, some not鈥攚hen they have a settled team and a comfortable style of play. Introducing a disrupting influence, by asking the team to play in a different way or bringing in a new player, for example, will push the team into a second phase鈥攁 period of chaos from which the outcome is unsure. Eventually, over a series of matches, the team settles into a third phase, a new equilibrium which may or may not be more successful than the first.
Lyons is trying to identify the types of perturbation and how to time them in order to lead a team through the chaotic phase into a third phase that is more successful than the first. 鈥淵ou only need to know that teams like Germany and Italy have remained successful over many years and through many changes in personnel to see that the introduction of new players and tactics can lead to an even more successful team,鈥� he says. 鈥淲e want to identify how these teams move through phase two to emerge stronger in phase three.鈥�
Without this information coaches may not risk upsetting their teams鈥� balance for fear of making matters worse. But if the team does not evolve it will ossify and coaches may store up problems for the future. Hughes points out that after England won the World Cup in 1966, subsequent teams modelled themselves, possibly unwittingly, on that champion team. And England have not won a major tournament since. 鈥淚f you ossify you are dead in the water,鈥� argues Lyons. 鈥淓ven if you have a patent for success it must be developed otherwise one day it will be superseded.鈥�
But deliberately moving from equilibrium into chaos can also produce startling results. In the 1970s, the Dutch introduced the concept of 鈥渢otal football鈥� in which players were encouraged to carry out many tasks, switching between positions during matches. The approach redefined roles and responsibilities, creating an entirely new style of play. It was a strategy that could have ended in spectacular failure but it succeeded and the team reached two consecutive World Cup Finals.
鈥淚nvesting in chaos may prove beneficial for teams if they have been relatively unsuccessful over a period of time,鈥� says Lyons. 鈥淧erhaps we need to introduce a culture where informed risk-taking is the norm otherwise systems never move into that second phase of chaos that allows evolution of a new team.鈥� Grehaigne agrees. He stresses, however, that introducing a perturbation should be left until the team has reached a period of equilibrium. If a team is not in a stable state, introducing change will not lead to phase two and could serve no purpose. 鈥淏ut when stability is reached you must change some players to avoid degeneration of the team,鈥� says Grehaigne.
Lyons and Hughes point to an example of an evolving chaotic system in last season鈥檚 English Premiership. Two-thirds of the way through the season, Newcastle United led the championship by a good margin and had games in hand over its rivals. Coach Kevin Keegan then bought the Colombian striker Faustino Asprilla. It was a high-risk strategy but Keegan recognised the potential the player would bring. But Asprilla鈥檚 arrival upset the team鈥檚 pattern of play, the team began to slip and eventually finished second to Manchester United.
Keegan may still have the last laugh, of course. His team may emerge from chaos into a more successful third phase. Keegan may have sacrificed short-term success for long-term gain. But from Newcastle鈥檚 record last season the Cardiff researchers draw another parallel with chaos theory. Even the tiniest tweak to a system can lead to massive and unpredictable changes in the future. In chaos theory this is called the butterfly effect: the argument goes that a butterfly flapping its wings in Sri Lanka can lead a few weeks later to a hurricane in Florida. In football the question is whether Keegan鈥檚 team would have been more successful if he had bought Asprilla a few weeks earlier, or eased him into the team more slowly?
Hughes and Lyons鈥攚ho admit their experience of international football is limited鈥攈ope to help coaches by taking some of the guesswork out of decision-making. 鈥淯sing data from notational analysis we aim to identify where decline sets in. This may be the time to introduce chaos and phase two鈥攖his is where you buy your Asprilla,鈥� says Lyons.
They also hope to be able to predict the results of changing a team鈥檚 tactics or players, both in terms of success on the pitch and over longer periods.
To some extent they can do this already. They can tell, for example, that certain tactics will not lead to absolute success鈥攁lthough they may result in an improvement. Ireland, for example, uses a tactic called 鈥渞eacher鈥� football which is designed to unsettle the opposition. As soon as the Irish players gain possession of the ball they kick it deep into their opponent鈥檚 half. Tall, powerful attackers then try to collect the dangerous bouncing ball and take the first opportunity to strike for goal. In other words, they mobilise perturbations caused by this style of play. This has brought Ireland relative success, allowing the team to win matches against more successful international sides.
However, reacher play has been promoted as a pattern for England which, because of its historical standing in international football, aspires to absolute success. According to Hughes, the reacher pattern will not work for England because one tactic is not enough for absolute success. 鈥淚f you merely apply this one method it becomes the norm, your equilibrium, and no longer upsets the opposition,鈥� he says.
Returning to this month鈥檚 European Championship, what lessons can the competing teams learn from chaos theory? Hughes argues that: 鈥淐oaches must be aware of all the skill variables that cause perturbations.鈥� These range from tackling in midfield coupled with intelligent passing and running, to the defender鈥檚 ability to disrupt the attack鈥檚 patterns of play.
Lyons has another message. 鈥淓ngland have worked hard to develop and refine a winning recipe for the European Championship,鈥� he says. 鈥淭here are enormous expectations placed upon the host nation in major tournaments. Going with what you know best is empowering but can lead to conservatism.鈥�
But Venables has shown that he is not tied to convention. In the match against Croatia in May, he introduced new players and changed the formation. By playing only three defenders rather than the usual four, with a very attacking midfield, England upset Croatia鈥檚 previously successful pattern of play and established new patterns and variables in its own play. This was a step into phase two. But can England exploit its newfound chaos in the next few weeks? 鈥淭he fascination of sport and chaos is that however close you come to manipulating and predicting the outcome you can never be 100 per cent sure,鈥� says Lyons. 鈥淭hat鈥檚 why both are so intriguing.鈥�
