By Simon Kuper and Stefan Szymanski
Published: April 16 2010 18:16 | Last updated: April 16 2010 18:16
When we began researching what would become our book Why England Lose & Other Curious Football Phenomena Explained, we decided not to believe a word that anyone said about the game. Instead we would test its shibboleths against data.
It was about time, too. For decades, football had escaped the Enlightenment. Clubs are mostly run by people who ignore data and do what they do because they have always done it that way. These people used to “know” that black players “lacked bottle”, and they would therefore overpay for mediocre white players. Today, they discriminate against black managers, buy the wrong players and then let those players take penalties the wrong way.
As Jean Pierre Meersseman, the cigarette-puffing Belgian director of AC Milan’s legendary Milan Lab, told us: “You can drive a car without a dashboard, without any information – and that’s what’s happening in soccer. There are excellent drivers, excellent cars, but if you have your dashboard, it makes it just a little bit easier. I wonder why people don’t want more information.” We did.
Not being ones to miss a trick, we’ve spent the past few months updating our facts and figures, ready for the World Cup this summer. Seeking insights about the tournament, we assembled new stats and tested old truisms. In England, we found that – for once – the pre-tournament hype might not be ridiculous: Fabio Capello’s team appears to be the country’s best side ever. But the main thing Capello needs to take to South Africa is probably a game theorist with a penalty database . And when Capello leaves, the Football Association should hire another foreign manager . And yet the likeliest world champions are not England but Spain – largely thanks, we learnt, to the country’s escape from fascism.
Some people may not want their emotions about the World Cup sullied by our rational calculations. On the other hand, if England lose a penalty shoot-out in the quarter-final, these same people will probably be throwing their beers at the TV, when instead they could be tempering their disappointment with some reflections on binomial probability theory.
Here is what to expect come summer.
The best-ever England team
There is a manic-depressive quality to supporting England. The nation tends to feel either very up or very down about the team. The night in November 2007 that Croatia kept England out of Euro 2008 was a low. But then Capello’s team ate up their qualifying group for the World Cup. Now English fans are dreaming about July 11, when their side could walk out for the final … Is the England team’s reputation as serial losers about to be rendered obsolete? And, if so, is there a Capello effect?
We believe that in the modern era, before the Italian became manager in 2007, England were always more or less equally good. That’s correct: the side’s strength barely changed over time. (This should have made the entire apparatus of punditry attached to the team instantly redundant.) A star player might fade or retire, but there was always someone coming up who was near enough his level to make the change almost imperceptible.
This may sound hard to credit. Fans feel strongly about the qualities of managers and players. There have been periods of national optimism and national pessimism, associated with the view that the England team are either strong or disgraceful. But, in fact, watching England play has resembled watching a coin-tossing competition. From 1981 to 2007, England on average won just over half their games; the rest they either drew or lost. So just as a coin has half a chance of landing heads, and half of landing tails, England in the average game had about half a chance of winning and half of not winning.
We assigned a “1” for each win and “0” for a loss or a draw, and examined the sequence of England’s 400 games from 1981 to 2007. Our finding: England’s win sequence over these games was indistinguishable from a random series of coin tosses. There is no relationship between a current coin toss and the last one: if you toss a fair coin there is always a 50/50 chance of heads. Similarly, there was no predictive value in the outcome of England’s last game, or indeed in any combination of England’s recent games. Whatever happened in the last match had no bearing on what would happen in the next one. All you could predict was that, over the medium to long term, England would win about half their games.
To make sure our finding was right, we constructed a few random sequences of ones and zeros to see if they looked like England results. Often we found more apparent correlation in our random sequences than in England’s results.
Fans and media have sought to see patterns where none exist. The best explanation for England’s ups and downs before Capello was randomness. But then the Italian arrived. He has now been in power for 22 matches, too short a period to draw firm conclusions, yet his England do seem to be better than any previous England team. Let’s take a deep breath, decide that two years without a major tournament is long enough to judge a manager, and compare his record with the entire terms of all other England managers. It looks good.
Most obviously, Capello’s winning percentage of 77 (counting draws as half a win) beats everyone else’s. His team also score freely: they defeated their opponents by an average of more than two goals a game, about a goal better than the typical modern England manager. Scoring was easier for prewar England sides, and for the first postwar manager Walter Winterbottom, because back then many international sides were terrible: in 1947, for instance, England won 10-0 in Portugal. If Capello’s goal difference is a guide, England could look quite convincing in the group stages in South Africa.
Yet beating minnows is rarely England’s problem. The trouble comes against the big teams, and Capello hasn’t changed that. His four defeats have been against France, Spain, Ukraine and Brazil. Losing to countries like that is no shame, but it won’t end 44 years of hurt either.
The English cannot manage themselves
Capello is an impressive man. He is the only England manager who has had an outstanding pedigree in club football: nine championships in 15 seasons. However, he succeeds with England not simply because of his personal genius. Partly, it’s because of his passport.
Our numbers suggest that the Football Association should keep upsetting xenophobes by importing foreign knowledge. From 1990 to 2010 England have had two foreign managers (Capello and Sven Goran Eriksson) and seven English ones, including two caretakers. On every measure, the foreigners have done better. It’s the same in major tournaments: Eriksson’s England never lost at the group stages and reached three quarter-finals, whereas English managers achieved one semi (at home) and one appearance in the last 16.
Eriksson and Capello outperformed with England not merely because they had a “golden generation” of players. Many men who played well for the foreigners were abysmal under Steve McClaren and Kevin Keegan. It’s painful to imagine what England might have achieved if the FA had stopped discriminating against foreigners decades earlier.
That a foreign passport makes a difference may sound odd. After all, these days England’s players have enough foreign experience. They play with, for and against foreigners every week. Why they need foreign managers was best demonstrated that night against Croatia. A team of Englishmen led by an Englishman played like caricatures of Englishmen. Steven Gerrard charged around in the rain, hitting impossibly ambitious passes, constantly losing the ball. He played like a headless chicken, or like an English footballer circa 1988.
For Liverpool, Gerrard usually plays like a European sophisticate. He has mastered the international, predominantly continental European style of top-class modern football. Against Croatia, however, there wasn’t a foreigner on the field or the bench to check him. When things began going badly, he shed that cosmopolitan skin, returned to childhood and played like an Englishman of old. He and his team-mates had spent their entire football education until the age of about 20 in England, playing an only slightly diluted version of traditional English football. Because of that upbringing, national football cultures continue to exist. That’s why England still need continental European managers to correct their flaws.
We also argue that most club managers barely affect results and could be replaced by their secretaries or by stuffed teddy bears without their team’s performance suffering. But international football is different. A manager can outperform when he brings to his team foreign knowledge that they didn’t previously have. Capello has done that for England, as has the Dutchman Guus Hiddink for the countries he has managed. Crucially, it’s easier to bring new know-how to a national team, where all the players lack the same foreign knowledge. Such knowledge gaps are much rarer in the English Premier League. All serious English clubs are now stuffed with foreigners, and know current international best practice. The Premier League is like an international market with almost perfect information. Consequently, managers scarcely matter.
He doesn’t always hit it to the right
Capello must solve one great problem: penalty shoot-outs. England have exited five of their last eight major tournaments on penalties. It remains hard to see them winning the World Cup without settling that bogeyman.
Most defeated England managers regarded penalties with fatalism. Glenn Hoddle, whose team lost in a shoot-out to Argentina in 1998, hadn’t even made his players practise penalties. A decade later, Hoddle admitted he should have trained his men in visualisation – visualising a penalty hitting the net, for instance. “Nine times out of 10,” he said, “if you miss, it’s between the walk from the halfway line … What I’m saying is, it’s in your mind where you’ve missed it … Nine times out of 10, you can tell if a player’s gonna miss it. With how he reacts, what his eyes are doing, when he gets the ball on the spot.”
Perhaps visualisation would have helped England. But a proper database of penalties might have been a better place to start. Some teams have tried to compile databases on how opponents take spot-kicks. Usually, the premise is that every penalty-taker has a favourite corner of the net and a set way of taking penalties. For instance, during Germany’s penalty shoot-out against Argentina at the last World Cup, the German keeper Jens Lehmann had a note tucked into his sock stating the supposed preferences of several Argentinians.
In fact, such cribsheets rest on a false premise. No regular penalty-taker always takes penalties the same way, because if he did, opponents would figure it out pretty quickly. Instead, penalty-takers try to vary their approach randomly. But most do have subtle recurring patterns that emerge under close analysis. One analyst is the Basque economist Ignacio Palacios-Huerta, professor at the London School of Economics. He got interested in penalties as a real-life test of game theory. Over time, he assembled a database on the habits of penalty-takers and goalkeepers. (Important disclosure: we are setting up a football consultancy. One of its aims is to bring Palacios-Huerta’s database to clubs.) Before the Manchester United vs Chelsea Champions League final of 2008, Palacios-Huerta sent a report about United and penalties to Chelsea’s then-manager Avram Grant. The report highlighted that:
1. United’s goalkeeper Edwin van der Sar dived to the kicker’s “natural side” too often. This meant that when facing a right-footed kicker, Van der Sar usually dived to his own right, and when facing a left-footed kicker, to his own left. So Chelsea’s right-footed penalty-takers might do better shooting to Van der Sar’s left.
2. United’s Cristiano Ronaldo “often stops in the run up to the ball. If he stops, he is likely (85 per cent) to kick to the right hand side of the goalkeeper.”
3. The team that wins the toss before the shoot-out chooses whether or not to go first. But this is a no-brainer: it should always go first. Teams going first win 60 per cent of shoot-outs, presumably because of the pressure on the team going second, which must always score to save the game.
The final went to a shoot-out. Ronaldo stopped during his run-up, kicked to the keeper’s right, and had his shot saved. Van der Sar dived too often to his right, and Chelsea’s first six penalty takers all kicked to his left. Chelsea would have won had John Terry not slipped taking his crucial kick.
For England to stand a chance of doing well, Capello must bring to South Africa a serious database on penalty-takers and goalkeepers – both the opposition’s and England’s.
The best team at the World Cup
From 1970 to 2000, a few continental European countries – Italy, Germany, France and Holland – worked out the best collective style of football. Each of these countries has its own preferences, but all share certain elements: fast, physical, collectivist, one-touch football. Their advantage was sitting in the most interconnected region in history. Football thinkers such as Arsene Wenger and Arrigo Sacchi could travel across porous borders, gathering and spreading knowledge.
From 1970 to 2000, the national teams of these countries piled up trophies. In the same period, the countries on Europe’s margins – the Brits, Iberians, former Soviet republics and Balkans – won none. They were isolated, excluded from the best knowledge networks and, therefore, stuck with their dysfunctional indigenous football styles. The Brits played kick-and-rush. The Greeks dribbled too much. However, from about 2000, the marginal countries came in from the cold. They became more integrated with core Europe, through travel, trade and football’s growing Champions League. Many countries – such as England and Greece – hired continental European football managers. Quickly, they absorbed continental know-how. Of all the formerly marginal countries, none did better out of this trend than Spain.
Fascism had isolated the country. “Europe stops at the Pyrenees”, was the saying at the time. Jimmy Burns, the Anglo-Spanish writer (and former FT journalist) born in Madrid in 1953, recalls in his book When Beckham Went to Spain: “Spain was virtually a closed economy. I spent part of my childhood between England and Spain, smuggling things from London to Madrid, never the other way round – clothes, gramophone records, books and magazines. While England seemed very much part of the world, Spain even to my young eyes struck me as something of a world of its own, where kids of my age all seemed to be taught by either priests or nuns.”
Spanish isolation was reflected in football results. The country won the European Championship at home in 1964, in the era before western Europe made its leap in football know-how, but then disappointed internationally for more than 40 years. What saved Spanish football was the country’s opening to the world. This began before General Franco’s death in 1975; two years earlier, Spain had reopened its borders to foreign footballers. It had also begun a long economic rise; contrary to the popular belief that footballers from poor countries are “hungrier”, teams from richer countries generally do better.
In 1986, Spain joined the European Union. Among many other things, this was a symbolic entry into Europe’s knowledge network. That soon paid off in football victories. Spain’s international results, decade by decade show an interesting story. The “winning percentage” column is very telling. It’s the best measure of Spain’s footballing performance per decade. It shows how closely Spain’s winning percentage tracks the country’s integration with Europe. In the 1920s, before the civil war, Spain excelled. But then isolation descended. From the 1930s through to the 1980s, Spain’s total winning percentage hovered around a disappointing 60 per cent. The European Championship of 1964 was an anomaly. The broader story was that a shut-off Spain struggled to access the best international football know-how.
But when Spain became more integrated, its football team got better. In the 2000s, Spain had a winning percentage of 81 (counting a draw as half a win), a performance about as good as any other national team has ever achieved. It became almost inevitable that the new Spain would win a trophy; if it hadn’t happened at Euro 2008, it probably wouldn’t have taken much longer. The Spaniards may not win this World Cup – the random element is high in such a short tournament – but they are the best team going there. Capello is pre-eminent among England managers. But we believe his England will return home in tears, like almost all their predecessors.
