Welcome to the Science of Sport, where we bring you the second, third, and fourth level of analysis you will not find anywhere else.

Be it doping in sport, hot topics like Caster Semenya or Oscar Pistorius, or the dehydration myth, we try to translate the science behind sports and sports performance.

Consider a donation if you like what you see here!

Did you know?
We published The Runner's Body in May 2009. With an average 4.4/5 stars on Amazon.com, it has been receiving positive reviews from runners and non-runners alike.

Available for the Kindle and also in the traditional paper back. It will make a great gift for the runners you know, and helps support our work here on The Science of Sport.

Wednesday, June 09, 2010

World Cup 2010 - Soccernomics 101

The formula for World Cup success

T-minus two days to the opening match (RSA vs. MEX), so best we start getting stuck into the football around here!  ESPN will broadcast all 64 matches on either ESPN, ESPN2, or ABC, and in case you have not checked just yet, the opening match will be 9:30 AM EDT this Friday, 11 June.  Where will you watch?

But to kick off our coverage, I heard an excellent interview yesterday with Simon Kuper, a transplanted South African and sports columnist for the Financial Times and co-author of Soccernomics:  Why England Loses, Why Germany wins, and Why the U.S., Japan, Australia, Turkey---and even Iraq---are Destined to Become the Kings of the World's Most Popular Sport.  Simon and his co-author Stefan Szymanski have analyzed just about every angle of world football and produced the "Freakonomics" text on the sport.  I have not yet read the book (but have ordered it already!), but from the interview here are some of most interesting points.  (Download the interview here.)

The first important finding. . .wealth = success

One of the ways people have tried to explain the East African dominance in running is that, beyond any physiological differences, those runners are "hungrier."  They come from poor backgrounds, very modest if not outright poor homes, and therefore are much more motivated to succeed.  When I ran the Bolder Boulder way back in 1998, the newspaper in Denver ran a story about how the Africans dominate the sport, and Frank Shorter chalked it up to incentive and more motivation, concluding that in one good year on the running circuit, the Kenyans can earn enough money to buy enough land and cattle back home to retire.  Right or wrong, this has been one angle to explain their success, or at least it is put forward as a contributing factor. 

So the "myth" is that countries with more poverty produce the better athletes because they are hungrier break through and work hard.  But perhaps not surprisingly, Kuper and Szymanski found just the opposite, that in fact poor nations are much less successful.  The higher per capita income, the better that country does in all sport.  Therefore wealth is a good predictor of success, as is the case with the Olympics, with Brazil the only developing country that is a world champion.

But wait, aren't the good players from poor backgrounds?  Indeed, yes, but in relative terms these poor players are still better off, and that means although they are "poor" by European standards, they are still better off than most of the world.  The explanation put forward is that since these kids grow up in small houses, without many options for leisure time, these future stars spend many hours playing their chosen sport.  So the best players come from poor backgrounds but from rich countries.

The socio-economic class of players

The authors gathered, for as many English players as they could, the professions of their fathers.  The finding?  The majority worked as manual laborers or in the football industry, so they had lower socio-economic status.  As Kuper outlines, in English soccer it is thought that if you are to be successful, you must leave school at 16 and pursue the sport full time.  That is probably a good thing if you do want to excel, because it will help ensure you maximize the time spent in the sport and in theory maximize your improvements.  As a rule, middle and upper class families encourage their children to stay in school and pursue higher education, even if the child demonstrates some aptitude at sport.  So England, at least, recruit their footballers from the working class population.  This is in contrast to countries like Holland, Argentina, and even Brazil, who are producing more top players from middle class backgrounds---apparently the Brazilian Kaka and the former Argentine star Redondo are from comfortable middle class families.

The problem with this is that in most first world countries, the working class populations are shrinking, and over time this means smaller and smaller talent pool from which to select your players.  So this "theory" states that eventually, unless England change something else, and given that the other countries continue on their paths, England's relative standing in the football world will decline.  Come back in 30 years for the follow up post!

It's not the coach, it's the players, stupid!

Many sports place coaches and managers on high pedestals, and often pay exorbitant salaries for their coaches.  To be sure the coach will always play an important role, and there are better coaches and worse coaches---even given the best players available, dare I say it but I could not coach that team to a World Cup final!  But what Kuper and Szymanski found was that a team's payroll is more likely to tell you how successful a team will be than who their coach is.  It is all about the players, because the better players demand higher pay, and the more money a club can spend on players means they can hoard more top players, therefore helping ensure their success.

The analysis was something like this. . .in any one season a club's payroll could explain 70% of how they finished in the league.  That's fine, because in any given season luck counts, and players get injured, etc., and cause even the richest clubs to under-perform.  However when averaged over a much longer period of say 20 years, the payroll accounts for 90% of the club's league standing.  On a statistical note, Kuper did not mention if those values were the "R" or the "r-squared" values, but either way payroll is a big determinant of success.  This finding holds true in both England and Italy.

So who will perform well in WC 2010?

Apparently three main factors predict success in World Cups:  1) a country's wealth, 2) population size, and 3) "football experience," i.e. how long they have been playing internationally.    Given these three items, Western European countries have the edge. However. . .historically these same teams perform less well when playing outside of Europe, so although the Western Europeans get the nod to go far into the tournament, they might do less well than in other years.

One problem with the World Cup, however, is that although it is one month in duration, the match ups that determine the latter match ups 0(2nd round and on) are decided over just three matches. . .and as Kuper mentions, just about anything can happen in three matches!  So there is a big element of randomness in the tournament that makes it harder to predict who will do well.

Interestingly, Spain gets the nod from Kuper this year because according to him although Spain has not won a World Cup ever, Spain is in fact a "soccer superpower" alongside Brazil.  This is based on their performances since 1990 at all levels of the sport, and even tough they do not have such a large population, they have (or had until the recession!) a high per capita income and also plenty of international experience.  Spain is probably a safe bet, and I think many people might also pick Spain to win it all this year as one can argue that they have the most complete team of all.

Applying Game Theory to penalties

As we mentioned in the "Intro" post earlier this week, penalties are a big part of the game because so many of the matches do go to extra time and end in penalties.  So an important part of the game is being successful in a penalty shootout.  Game Theory fits in here because it in a penalty shootout we have two players who actions are dependent on each other.  The shooter will make a decision to kick the ball one way or the other, and based on what the keeper thinks that decision will be, he then makes a decision about which way to dive.  In fact way back in 2002 Steven Levitt co-authored a paper in The American Economic Review in which they tested the predictions of game theory using penalties.  (Read Levitt's Freakonomics post about it here.)

In the interview Kuper talks about how Chelsea in 2008 hired an economist to advise them on their penalties strategy for the Champions League final against Manchester United.  Apparently the United keeper, Edwin Van der Sar, dives to his right way too often. . .and so it goes that the best shot against him is to his left.  Accordingly, the first five shots by Chelsea were kicked to Van der Sar's left, all scoring except when John Terry hit the post on---where else?---Van der Sar's left. 

Kuper deferred to the video on Youtube, and we will do the same here, and you have to watch it, because on the second kick in sudden death, Van der Sar points to his left. . .signalling to the kicker, Nicholas Anelka, "I know you are going to kick it left."  It was the ultimate second guessing game, because Van der Sar was able to recognize the pattern of Chelsea's strategy, called Anelka on it, made him second guess his decision to kick left, thus "forcing" Anelka to kick right, and in the end saved the kick and won the match for his team.

Needless to say, I can't wait for the book to arrive and am really looking forward to reading it in its entirely.  It is a fascinating analysis of the sport, with some results perhaps not so surprising---wealth produces success---but others most surprising, in this case the socio-economic status of most English players.  Whether you are watching on the internet at work or over breakfast with friends, or at sunrise on the west coast, polish up you vuvuzela and enjoy the first match.  Go Bafana!



Gus said...

The book is interesting but has a lot of flawed arguments which can be torn apart easily - it is no Moneyball.

My money is on Brazil - good as Spain are I think Xavi will tire badly in the tournament, and he is the key to Spain.

Joe Garland said...

History of soccer in the US: We coulda been a contender.

Oliver said...

Haven't read the book but it is not as cut and dried as that.

Poor people from poor countries do well at sport that has 'simplicity'and requires little 'apparatus' etc.

Distance Running: They (poor people)will continue to do well in that sport in their own country if that sport requires little infrastructure for competition and that sport has little competition from other sports that are more 'attractive' and with more probability of financial success...eg within eastern Africa.

Soccer/football as an example: Again, a very easy sport to take up as a kid, lots of fun etc. Therefore lots of good players ex Africa, South America etc. I remember as a kid in SA we played football every day after school on a vacant lot, or in the street. Anything that resembled a bll was used.
However, at organsised club level it requires finances and structure for success, hence the domestic programs in poor countries may not be good.

Financially strong countries:
Yes, England has done well and will continue to do so, because there is a history and love for the game.
The game within the country is played at a high level because there is no salary cap and hence the good players who have persisted in the poor countries can make money there and elsewhere.
Countries such as Australia 'punch above their weight' by getting lots out of the few that re elite through the money and structure, but it is in 'western' sports where the 'pool is shallow' eg swimming...high participation rate, little competition, lots of structure and coaching or sporst such as hockey, tennis etc

When it comes to soccer/football however, Australia does not have either the 'culture', 'participation rate', structure nor is it in a 'shallow pool', hence the incusion of Australia in that group is a bit wishfull. They will always be there or there abouts but not the best.

Money alone won't do it. They won't be able to replicate an "english league" to keep good players and attract migrant players and history has shown that even transplanting a sport of little following into an area (eg Rugby league in Melbourne) doesn't create the sport within that area.
The sport is also still dominated by European migrants in a country that is dominantly anglo and Rugby league/AFL.
Kids at school will dabble in running/soccer etc but the best can make more money out of league, AFL, basketball etc. Good swimmers will remain in the pool and most Aussie kids can look forward to a descent education/career so if you are undecided about professional sport, you can make a more comfortable immediate lifestyle by a few years study and a career, which is another competitor to sports in western countries.

Every four years the Aussies back the socceroos until they get knocked out of tournament and in between it is just the hard core that follow the sport.
They have one of the richest people in world as CEO, but that isn't enough to develop a culture...sorry.

The poor countries, Africa etc will keep producing more and more good players ho no realise they can make great money overseas (think of the numbers that hve increased since the 80's), however their domestic competitions will never be on par.

However, by sheer weight of numbers tghey will eventually harness that individual talent into good teams with good coaching.

Lastly , just a comment on the coaching aspect...another area where Australia has shown that even 'journeymen' (because that's all that Neil, Kewell, Emerton et al) can get to and succeed to certain level at WC through a proven coach (Hiddink, Verbeek etc)

This may happen in Africa too.

Sorry for long disjointed post, just wanted to get it of in hurry.


Gene said...

I think it's important not to confuse good marketing with good science. Take a look at "freakonomics" outside of sports for a start, to get a better feel for their methods and arguments, and the critiques. I quit reading the freakonomics pieces in the NY Times because they were good ole' 19th century rational utility maximization-based arguments given new life with some econometric computer fishing and cutsey writing. Sorry, but studies regularly show the world doesn't work that way (nor do I, and I bet you don't either).

In any case, the 0.9 they claim has to be R (correlation), not R-squared, the amount of variance explained. Virtually nothing in this world can be accounted for at that level. Now, an R-squared of 0.3 will go far in academia, but generally doesn't translate well to the real world.

Sports are a good example of why not: just think about the baseball NY Yankees, how much they spent over the last decade and how little they got for it, until they changed the talent and personality chemistry of the team. In the meantime, relatively low-budget owners in smaller markets were among those winning titles with chemistry and enough talent. In other words, there's a heck of lot more that goes into being a champion than the amount of money one has, whether it's individual money, team money or the country's money.

Anonymous said...

You might enjoy this from the NY Times, "How a soccer star is made"

Interesting that the majority of players at the Ajax academy are middle class and that the scouts follow 5-year olds. Also, the academy emphasizes developing individual skills over winning games in stark contrast to some coaches (US) who want to always win. Mike