Friday 16 December 2011

[Jonathan Wilson] The Question: just how competitive is the Premier League?

Based on a study of world leagues, it may be that England's top flight is not the fierce contest its branding suggests
To read part two of this series, click here
Premier League: Manchester United v Norwich City
Norwich impressed at Old Trafford and showed the kind of spirit that is a Premier League selling point. They lost 2-0, however.

It's the most competitive league in the world. Anybody can beat anybody on their day. There are no easy games in this league. The mantra of the Premier League apologists is well known. Every time a team from the Little Fourteen (as nobody ever calls them) plays a team from the Big Three (it is just three now, right?) the cliches come trotting along: in our league nobody gives anybody anything, everything's a glorious struggle.

It's nonsense, of course: it's obvious the Premier League is a closed shop that can be opened only with the application of around a quarter of a billion pounds. In England in the past decade there have been three different champions. That's the same as Spain, Italy and Portugal, poorer than France (four), and Germany and Russia (five). But then in the past decade there have been Champions League winners from England, Spain, Italy and Portugal, and not from France, Germany or Russia. It seems fairly obvious that competitiveness is something that must be balanced against quality: would fans prefer an exciting domestic league, or for teams from that league to do well in continental competition?

That's not to say that a low number of different champions is necessarily a sign of strength or high quality. In Croatia, Dinamo Zagreb have won the league for the past six seasons and made next to no impression in Europe. The problem, Igor Biscan said, is that their players get used to winning easily; come a tough match, a game against a side of even slightly lesser ability, in which they have to do things that don't come naturally, like defending, they have no idea what to do. "Our players walk through games against villages so they forget how to run," as a Crvena Zvezda director put it to me a couple of years ago. "But what are we meant to do? Buy players for the other teams in the league as well?"

Rangers and Celtic have perhaps suffered from that at times in Europe, which may mean that the domination of European football by Barcelona and Real Madrid many have predicted is not quite so sustainable as many think. That in turn is something to consider for those who would tweak the balance of competition in the Premier League by doing away with collective bargaining for TV rights (quite apart from asking whether the product will remain so appealing if games are hideously one-sided). That answer to the Zvezda official's question, in fact, may end up being "yes", if only indirectly.

If you want a true range of champions, you need to leave Europe. There have been six different champions in Japan in the past decade, seven in Brazil. In Argentina, where the apertura-clausura system means there have been 20 championships in the past decade, there have been 11 different winners – but there the spread of champions seems a function of weakness, with the best players from the best sides being skimmed off by predators from Europe and Brazil after each championship in what's effectively a reverse of the draft system in US sports.

But even if we accept competitiveness per se as a good thing, there are different types of competitiveness. After all, while there have been five different champions in the past decade in Germany, Bayern Munich have won the title five times, the same number as Manchester United, Barcelona and Internazionale (Porto and Lyon, incidentally, are the most successful individual clubs in the 10 leagues considered with seven titles each in the past decade). In effect, in Germany there is a Big One and, if they fire, nobody else has much of a chance. Whether one giant and a handful of occasional challengers is preferable to two or three giants is debatable.

Looking at the number of champions, though, says little about whether a team at the bottom can beat a team at the top. To try to come up with a statistical basis for assessing competitiveness within a league, I looked at four metrics across the 10 leagues (England, Spain, Italy, Germany, France, Portugal, Russia, Brazil, Argentina and Japan) over the past decade: the average gap from first to second at the end of the season (how dominant is the champion?); the average gap from first to fourth (is it only two or three teams who challenge the champion?); the average gap from first to last (what's the gulf in quality from top to bottom?); and the average gap from fourth to fourth-bottom (what's the difference in quality between the mid-ranking sides?). Because different leagues have different formats and are different sizes, these gaps have all been expressed as points-per-game. (Points deductions were ignored). 

Team
1st-2nd
1st-4th
1st-last
4th-4th last
England 0.17 0.52 1.92 0.79
Spain 0.14 0.5 1.42 0.62
Italy 0.19 0.53 1.66 0.68
Germany 0.2 0.4 1.4 0.71
France 0.19 0.42 1.3 0.54
Portugal 0.27 0.72 1.68 0.64
Russia 0.11 0.34 1.49 0.68
Brazil 0.17 0.33 1.18 0.56
Argentina 0.17 0.46 1.5 0.7
Japan 0.19 0.36 1.39 0.62

In terms of exciting title races, it turns out that Russia is the place to be, with Spain some way back, and the rest trailing far behind. Perhaps not surprisingly, given Porto's domination, Portugal has the highest gap from first to second, but what is striking is that Germany has had the second-least-close title races over the past decade, a fifth of a point-per-game separating first from second. The gap between first and fourth in the Bundesliga, though, is the fourth smallest of the 10 leagues surveyed, the gap from top to bottom the third smallest and the gap from fourth to fourth-bottom the smallest. That suggests that the leader often streaks away while the rest remain relatively tightly bunched.

The biggest gap from first to fourth is in Portugal. Again, that's in line with expectations: since the second world war, only Belenenses and Boavista, once each, have interrupted the flow of titles for Porto, Benfica and Sporting. There is a very clear historical Big Three who continue to dominate. In Italy, similarly, the two Milan clubs and Juventus (calciopoli notwithstanding) have clearly been dominant. What is perhaps unexpected, though, is that England have the third-biggest average gap from first to fourth: perhaps the Big Four was always something of a myth.

Brazil has the smallest gap from first to fourth, which it is tempting to ascribe to the size of the country. A population of almost 200 million can perhaps sustain more big clubs that smaller nations. It is also worth noting, though, the relative immaturity of a national championship in Brazil; it could be that as the present system becomes more established, money and success gravitates to a more select few.

But what's really telling is the last two columns, which show that there is a bigger gap between top and bottom of the Premier League and between fourth and fourth bottom in the Premier League than in any of the other nine leagues under consideration. Far from being the most competitive league in the world, in fact, it turns out to be the least. The league where the bottom is closest to the top, rather, is Brazil, which is remarkable when you consider that the statistics include the 28-team top flight of 2001 and the slow contraction to 20 in 2006 – the more teams there are, the wider you would expect that divide to be.

Now of course an analysis of points tells only part of the story. It may be, in some hard-to-quantify way, that lower Premier League teams fight harder before losing to the big guns, and it certainly is true that the culture of arranged games, mutually beneficial draws and the like, seems less pronounced in England than elsewhere.

In terms of hard statistics, though, the message is clear. The Premier League may lead the way in terms of marketing and self-promotion, but if you want competitiveness, go to Russia or Brazil.

Next week, I'll look at how competitiveness has changed in England over time, what the reasons for those changes may be and what the potential impact of scrapping collective bargaining for television rights may be.

No comments: