Soccernomics (7 page)

W H Y E N G L A N D L O S E S A N D O T H E R S W I N

39

Now let’s look only at matches between European countries—the arena where England plays most of its soccer. The match results in Russell’s database show that home advantage and experience count for a little less in Europe than in the world in general. On the other hand, population and GDP count for a bit more in Europe than in the world.

Taking all that into account, against European teams England overperforms very slightly. From 1980 through 2001 it scored a goal every twenty games more than expected given its experience, income, and population. That put England in 23rd place of the 49 European countries we ranked. It was just behind West Germany in 20th place, but ahead of France (25th) and Italy (31st), even though those countries won tournaments in this period and England didn’t.

Our conclusion: England does just fine. The team actually performs better than expected, given what it has to start with. All it needs to bring home some trophies is better timing—it must win fewer friendlies and more World Cup semis—and a few million more inhabitants.

Consider the England-Germany semifinal at Euro ’96. Both countries had similar levels of experience, and by 1996 Germany’s lead over Britain in income per capita had slipped to only about 7 percent. However, according to our analysis, England’s home advantage (worth just under half a goal in a game between Europeans) was largely wiped out by Germany’s much larger population. That made the expected goal difference only 0.26 in favor of the English home team, not far short of the tie that materialized in open play. In the end Gareth Southgate missed his penalty, and England lost.

Overall in the 1980–2001 period, England played Germany ten times. Taking the results at the end of normal time, England won two, tied three, and lost five. Its goal difference over the ten games was minus 2. Given Germany’s slight lead in GDP per head in these years, but particularly given England’s shortage of people, that is almost exactly the goal difference we would have predicted. The “thirty years of hurt” shouldn’t be a mystery.

40

PHASE 8: ENGLAND ENTERS THE NEXT WORLD CUP

THINKING IT WILL WIN IT

Famously, there is a manic-depressive quality to supporting England.

The nation tends to feel either very high or very low about the team.

The night of that home defeat to Croatia in November 2007 was a low.

As we write, in summer 2009, most English fans feel high again.

Capello’s team is eating up its qualifying group. People are starting to daydream about July 11, 2010, when a certain side might just be walk-ing out for the World Cup final in Soccer City outside Johannesburg.

It sounds reasonable to presume that a good qualifying performance presages a good tournament. But is this belief backed up by the facts?

When England qualifies in style, is it really more likely to excel at the final tournament? We checked the data.

England has tried to qualify for fourteen World Cups and Euros since 1980 (it didn’t have to qualify for Euro ’96, because it was the host). Of those fourteen tournaments, it failed to qualify for three, and got knocked out in the first round four times. That means that half the England teams in this period can be considered failures. Collectively, we’ll label these teams “Bad England.” Of the remaining seven tournaments, England reached the round of sixteen twice, the quarterfinals four times, and the semifinal once, in 1990. We’ll label these seven teams “Good England.”

You would assume that Good England did much better in qualifying matches than Bad England. After all, Bad England regularly failed even to qualify. Yet the data show otherwise: in qualifiers, Good En -

gland and Bad England were much of a muchness.

The England team at Euro ’88, for instance, was unmistakably Bad England. It lost all three games in West Germany before going home with the hooligans. Yet it had qualified gloriously: five wins, one tie, no losses, and a goal difference of nineteen for and just one against. By contrast, the Good England team of 1990 had muddled through qualifying, winning three and tying three.

Even more curiously, England often performed just as well in qualifiers when it failed to make the tournament as when it succeeded. For W H Y E N G L A N D L O S E S A N D O T H E R S W I N

41

example, in missing Euro 2008, England got exactly the same percentage of the available qualifying points (namely 66 percent) as when it managed to qualify for Euro 2000 or for the World Cup of 1982.

Figure 2.2 shows a comparison of the qualifying records of Good England (teams that made at least the round of sixteen) versus Bad England (teams that either missed the final tournament or went home after the first round). The figure shows that Good England and Bad England had almost indistinguishable win-loss records in qualifying.

Indeed, Bad England scored many more goals than Good England while conceding only slightly more. The lesson of history: England’s performance at specific World Cups and European championships is almost entirely unpredictable.

F I G U R E 2 . 2
England’s qualifying record since 1980

Team

Played

Won

Tied Lost

For

Against

Bad England

60

36

16

140

33

Good England

56

36

13

106

28

What we see here is partly the enormous role of luck in history. We tend to think with hindsight that a team that did well in a particular tournament was somehow always going to do well, and a team that lost was doomed to do so. The winner’s victory comes to seem inevitable.

This is a common flaw in the writing of any kind of history.

But in fact, inevitable victories hardly ever happen in soccer tournaments. Perhaps the only recent case was Brazil at the World Cup of 2002. Just how dominant that team was dawned on a leading European club manager a few months after the final. This manager was trying to sign Brazil’s goalkeeper, Marcos. After all, Marcos was a world champion. Marcos visited the club and did some physical tests, in which he didn’t perform particularly well. Never mind, thought the manager, the guy won the World Cup. So he offered Marcos a contract. At two o’clock the next morning, the manager was awoken at home by a phone call. It was Marcos’s agent.

The agent said, “I’m sorry, but Marcos won’t sign for you.”

The sleepy manager said, “All right, but why not?”

42

Then the agent confessed. A couple of years before the World Cup, Marcos had broken his wrist. It had never healed properly. But his old club manager, Luiz Felipe Scolari, became manager of Brazil and put Marcos on the team. Suddenly, Marcos was going to a World Cup.

Every day at the tournament, the agent explained, Marcos was in pain.

He could barely even train. In matches he could barely catch a ball.

Every day Marcos told himself, “I really must tell Scolari about my wrist.” But he could never quite bring himself to. So he went on, day after day, until he found that he had won the World Cup. Brazil was so superior that it became world champion with a crocked goalkeeper.

However, such dominance is very rare. Normally, the differences between teams in the final stages of a World Cup are tiny. The difference between an England team being considered legendary or a failure is two to three games, each generally decided by a single goal, over two years.

After all, the difference between making a World Cup and spending the summer on the beach can be just a point. Sometimes it’s a point that you lost by hitting the post. Sometimes it’s a point garnered by a rival in a match you didn’t even play in.

Once you’re at the World Cup, the difference between going home ignominiously in the first round and making the semifinals is often a matter of a few inches here or there on a couple of shots. The greatest prize in the sport hinges on a very few moments. It’s the same in baseball, notes Michael Lewis in
Moneyball
: “The season ends in a giant crapshoot. The play-offs frustrate rational management because, unlike the long regular season, they suffer from the sample size problem.”

Lewis means that because there are so few games in the play-offs—because the “sample size” is so small—random factors play an outsize role in determining the winner. It’s the same in soccer World Cups, and even more so in European championships. “Euros” last only three weeks, and as Arsène Wenger notes, any team in a league can be top of the table after three weeks.

Most fans understand that luck matters, even if they construct a postfact story about the tournament that makes either victory or humil-iation seem fated from the start. But our data point to an even scarier W H Y E N G L A N D L O S E S A N D O T H E R S W I N

43

truth than the existence of fluke: namely, there is barely any difference between “brilliant” and “terrible” England teams. It looks suspiciously as if England is always more or less equally good.

This may sound hard to believe. Fans feel strongly about the qualities of managers and players. There are periods of national optimism and national pessimism, associated with the view that the England team is either strong or disgraceful.

But in fact, watching England play resembles watching a coin-tossing competition. If we focus on outright victories, then England on average wins just over 50 percent of its games; the rest it either ties or loses. So just as a coin has half a chance of landing heads up and half tails up, England in the average game has about half a chance of winning and half of not winning.

We assigned a “1” for each win and a “0” for a loss or a tie, and examined the sequence of England’s four hundred games since 1980. Before we discuss this sequence, let’s look at coin tossing. If you tossed a coin four hundred times, you would expect on average to get two hundred heads and two hundred tails. However, there is no reason to think that the outcomes would alternate (heads, tails, heads, tails . . . ). Sometimes you will get sequences of a few heads, sometimes a few tails. Crucially, though, there would be no relationship at all between the current coin toss and the last one. If you toss a fair coin there is always a fifty-fifty chance of heads, whatever sequence has occurred up to this point. There is no statistical correlation between current coin tosses and past coin tosses, even if the average of any sequence is always around 50 percent.

Here is our finding: England’s win sequence over the four hundred games is indistinguishable from a random series of coin tosses. There is no predictive value in the outcome of England’s last game, or indeed in any combination of England’s recent games. Whatever happened in one match appears to have no bearing on what will happen in the next one.

The only thing you can predict is that over the medium to long term, England will win about half its games outright. We have seen that the outcome of matches can be largely predicted by a country’s population, income, and experience. However, this explains only the
average
outcome.

44

In other words, if England were smaller, poorer, or less experienced, it would have a lower percentage of wins, but the sequence of those wins would still be unpredictable.

To make sure that our finding was right, we constructed a few random sequences of 1s and 0s to see if they looked like England’s results.

Often we found more apparent correlation in our random sequences than in England’s results.

Contrary to popular opinion, it may be that the strength of the England team barely ever changes (which would make the entire appa-ratus of punditry attached to the team instantly redundant). A star player might fade or retire, but in a country of 51 million people, there is always someone coming up who is near enough his level to make almost no difference. Over the long term, the three key factors that determine a country’s performance are very stable. The British economy boomed in the 1990s and is now dropping down the global wealth rankings, but measured over the last century Britain has always been one of wealthiest nations in the world. Equally, its share of the soccer population changes only glacially. And while the England team gains experience, so do its main rivals. The only key factor that changes is home advantage. Given that playing at home is worth a lead of two-thirds of a goal per game in global soccer, it’s little wonder that England won the World Cup in England in 1966.

Otherwise, England’s performances in good or bad times are much the same. It’s just that fans and the media seek to see patterns where none exist. Nick Taleb, the financial investor who wrote
The Black
Swan: The Impact of the Highly Improbable
, famously explained that we are constantly fooled by randomness. In neuroscientific terms, our rational brains are egged on by our emotional brains to find patterns even if there aren’t any. In the end, the best explanation for the short-term ups and downs of the England team is randomness.

PA RT I

The Clubs

Racism, Stupidity, Bad Transfers,

Capital Cities, the Mirage of the NFL,

and What Actually Happened in

That Penalty Shoot-Out in Moscow

GENTLEMEN PREFER BLONDS

How to Avoid Silly Mistakes in

the Transfer Market

In 1983 AC Milan spotted a talented young black forward playing for Watford. The word is that the player Milan liked was John Barnes, and that it then confused him with his fellow black teammate Luther Blissett. Whatever the truth, the Italians ended up paying Watford a “transfer fee” of $1.4 million for Blissett.

As a player Blissett became such a joke in Italy that the name

“Luther Blissett” is now used as a pseudonym by groups of Italian anar-chist writers. He spent one unhappy year in Milan, before the club sold him back to Watford for just over half the sum it had paid for him. At least that year gave soccer one of its best quotes: “No matter how much money you have here,” Blissett lamented, “you can’t seem to get Rice Krispies.” More on Rice Krispies later.