Aug 8, 2012

How can a country win more Olympic medals?

Olympic watchers who aren’t happy with their country’s absolute medal count can always get solace from one of the many places offering alternative medal tables, like

Alan Davies — Editor of The Urbanist

Alan Davies

Editor of The Urbanist

Forecast and actual medals won by top 15 medal winners at Beijing Olympics 2008 (from Forrest et al)

Olympic watchers who aren’t happy with their country’s absolute medal count can always get solace from one of the many places offering alternative medal tables, like The Guardian’s Datablog.

It shows that at the end of day 12 New Zealand, for example, was lying in third place on the population-adjusted medals ladder. That’s much better than its official position (sixteenth) and, perhaps equally important from a Kiwi perspective, it’s better than Australia was doing at that stage!

But whatever consolation population-adjusted rankings might offer, population size has precious little to do with winning Olympic medals. Sure, the most populous nation on earth is currently leading the medal count in London, but the second most populous country is only ranked forty-sixth.

In a model they constructed to predict the share of medals won by countries at the Beijing Olympics, academics David Forrest, Ismael Sanz and J.D. Tena found that when GDP was included, population size had little direct role in explaining the relative performance of countries. What seems to matter far more are factors like national wealth and a country’s preparedness to direct resources to fostering elite sport.

Their model was built on six variables (since they’re closely related, they call them co-variates). For each country, they measured GDP; the number of medals won at the preceding Olympics; government spending on sport and recreation; and whether or not the country is the Olympics host; is due to host the next Games; was a member of the former Soviet bloc; or is a planned economy.

They calibrated the model using the medal results for 127 countries (those for which data was available) at each Olympic games from 1992 to 2004. They then forecast the total medals each would win at the 2008 Games and subsequently compared the prediction against the actual results from Beijing.

It’s evident the forecasts understated China’s, Australia’s and the US’s medal tally somewhat and overstated Japan’s (see exhibit). However the authors say the R2 between the forecast and actual medals was 0.970.

Moreover, 99 countries were within two medals of the forecast. That’s not surprising given most were in a long tail, but the model was successful in predicting thirteen of the countries that filled the fifteen leading positions and in broad terms it got the rank order of those 15 right.

So the ability and preparedness of countries to channel resources into elite sport seems to be the key determinant of Games success. Great Britain is a classic example. It won only 28 medals at the Sydney Games and 30 at Athens, in both cases well behind Australia’s total (58 and 49). However following London’s selection for the 2012 Games, GB’s tally jumped sharply to 47 at Beijing, one more than Australia, and it’s won 48 so far (end Day 12) at London.

According to Tyler Cowen and Kevin Grier, there are nevertheless other ways countries can increase their chances of winning more medals.

One strategy is to focus available funds on those sports which offer lots of medals. Team sports like hockey, soccer and basketball are expensive because many players have to be funded, but typically only offer a single medal.

In contrast, sports like swimming, athletics, shooting, judo and cycling offer many medals and only require one or a handful of competitors per medal event. In some sports, a single competitor can enter multiple events and reasonably expect to do well. Michael Phelps, for example, swam in seven events in London (winning five medals) and Ryan Lochte swam in six (winning five medals).

Another strategy is to focus on sports in which the country has a comparative advantage. Cowen and Grier cite the fit between beach volleyball and the “outdoorsy” lifestyle of California, Brazil and Australia. Perhaps a better example is the fit between Australia’s high incomes and sports that are expensive for participants, like equestrian and sailing.

Kenya has a comparative advantage in distance running. Jamaica excels in sprinting, but rather than being the result of a “natural” endowment, it appears to be a specialisation it’s created. Cowen and Grier also suggest countries keen to increase their medal count can apply available funding to “importing” talent from other countries.

Steve Sailer says rich countries have sought to increase the number of women’s events because it’s the “easy way” to win more medals.

If the Olympics still had the same distribution of events as in, say, 1952, poor countries would win a larger percentage of medals. Training women for some obscure macho sport is a luxury that only rich countries and dictatorships can do.

There’s little doubt many advanced countries, including Australia, already implement some or all of these principles (I ignore illegal methods). They do it because success at the Olympics really matters to lots of people in lots of countries. It doesn’t matter for instrumental reasons – it matters because it makes those citizens feel good.

If the citizens of Australia knew the real cost of medals they might possibly think twice about the level of public funding provided for Olympic sport, but I very much doubt it. Shaun Carney says the Australian Institute of Sport was established to buy Olympic gold medals and gets it right, I think, when he says,

there is not a shadow of a doubt that this had the support of a big majority of Australians. We like watching men and women in the Australian strip winning. It makes the rest of us feel like winners. In fact, the rest of us claim the victories as our own. ”We” won 16 gold medals in Sydney in 2000. ”We” took 17 gold at Athens in 2004.

Personally, I wish Australians were as passionate and proud about our fellow citizens achievements in fields like managing the environment, social justice, establishing new businesses and technological innovation, as we are about their sporting successes. Imagine if we saw addressing disability and fostering creativity (say) as grand national projects on a par with success at the Olympics.

It’s tempting to think that as we become more confident as a nation of our place and status in the world, we’ll be less inclined to focus on sport. But I think that’s unlikely. Countries with longer histories than ours, like the US and Great Britain, show little sign of pulling back on Olympics expenditure. Indeed, our Olympic officials (here and here) reckon a key reason why Australian athletes have won fewer medals at these games is because other nations have upped the expenditure ante.

P.S. Some researchers at Ruhr-Universität Bochum have made medal predictions for the London Games. Unfortunately I couldn’t find full details of their methods. They predict Australia will win 43 medals in 2012 and rank fifth in the total number of medals won (at the end of day 12 Australia had 26 medals).

(Visited 68 times, 1 visits today)


Leave a comment

Your email address will not be published. Required fields are marked *

10 thoughts on “How can a country win more Olympic medals?

  1. Alan Davies

    Scott #3, fisher greg #9

    Anyone who reads their paper will see Forrest et al are well aware of collinearity (as I mentioned in para 5 above).

    Yes, of course there’s some correlation between GDP and population but there’s a host of other variables that are also correlated with GDP (e.g number of sporting facilities), so researchers rightly focus on GDP ahead of any of them because it’s the underlying force.

  2. fisher greg

    Scott. While I broadly agree with your analysis, there is a protocol issue I would like to raise.
    I understand that there are rules that must be obeyed when commenting of statistics presented as bar graphs in online publications and newspapers related to both political and sporting matters. One of those is that you can not use the term multicollinearity, because it shows that ( in the case of Crikey) you have a good understanding of statistics (and in the case of the Sun Herald, you know a word of more than two syllables). The second is that you can not put the word “independent” in quotation marks when discussing “independent variables”, but rather need to put the whole phrase “independent variables” in quotation marks as it because it will produce more responses from people who have a pedantic view of grammar.

  3. michael r james

    Oops, should have read:

    It was notable on Gruen Sweat last night that both mock adverts to convince an audience that the $365 million we spent on supporting Olympic ambitions would be better spent on other more worthy recipients–ie. don’t give any money to the Olympic effort, were quite convincing.

  4. michael r james

    Hah! No more than seconds after posting my comment than Clark and Dawe (7.30) propose spending all the Mining SuperProfits Tax on Australian Sports because no Australian is going to object to that and the miners wouldn’t dare raise a fuss.

    See, it’s not so silly is it?

  5. michael r james


    Well, of course, the sporting fields of Eton built the Empire, the Commonwealth and won all that Olympic gold. BoJo (Boris Johnson) could win an Olympic gold medal for buffoonery.

    But actually I don’t know how many of those bloody Yorkshiremen went to Eton but a Brit newspaper examined the geographical distribution of medals and found that county not only led the country but also (until today perhaps) were beating Australia. Meh.
    I have been wondering about the Australian obsession and funding thing. It was notable on Gruen Sweat last night that both mock adverts to convince an audience that the $365 million we spent on supporting Olympic ambitions, were quite convincing. I think there is something to the joke Seb Coe (before the Games began) said it was clear why Australia so often beat the UK in sport: clearly they have no cultural distractions at all!

    How can I say that in the week one of our intellectual giants died? Because for the majority–and I fear a growing majority–of Australians won’t even know who Robert Hughes is, or what he is famous for, or have read anything he wrote, or even seen any of the groundbreaking docos he made. Take a poll amongst genXYZ that would be true and Seb Coe is not so far off.

    Incidentally I wonder if the US government spends anything like Australia (or China?) on Olympic ambitions? Other than indirectly (via supporting its universities) I doubt it. Anyone showing promise will get private sponsorship or university sponsorship (even tennis gets supported this way; it is the only reason why John McEnroe was at Stanford!) Most of their sporting stars come out of their university sporting schemes; and the only country that comes close to what the US has in universities in this regard is Australia. But we are well into a deliberate program to downgrade our universities into mere credentialization factories. And managerialism (which destroys everything it touches including higher-ed) is the death of almost anything worthwhile–and that is where this half a billion dollars of sport money goes.

  6. Tom the first and best

    There have been more Olympic victories by Eton graduates that Africans and Britain`s Olympic team has a higher proportion of elite background persons than the students of Oxbridge (at least I read that somewhere). The Olympics are full of sports that require expensive equipment and facilities even for entry level competitors that are also entirely leisure activities.

    It also invades the media with even more sport than normal. I do not like the Olympics.

  7. Last name First name

    What a silly question, the Olympics entertains the world and me since 1948 and I hope to to see the next one in Brazil hope fully before my use by date. I hope all country carry on with the Olympic spirit being paramount on as it always has for most countries. To me running was always a joy until I got wasted by a car. The olympics is about joy not number of medals. Ignore deadhead bean counters.

  8. Scott

    Using simple trend analysis from the last two games, you get Australia getting 43 medals….11 Gold, 14 Silver, 18 Bronze. Problem is, no one predicted Australia’s underperformance in the pool.
    As for the research listed above, there is bound to be some multicollinearity in the “independent” variariables. GDP and Recreation spend are bound to be correlated pretty heavily (which is the real reason why population was left out…GDP and population are heavily correlated so GDP is a good enough variant to capture the population effect).
    While this won’t affect the prediction, it will impact on the effect of each variant on medal count (and will increase the R2 value). It’s a pity that the research did not include a correlation matrix (which is pretty standard for statistical analysis).
    All up it appears that how you went in the last games is still the best predictor of future success.

  9. xasd

    is interested that you mentioned Australia’s achievement in addressing disability. Last I saw, Australia has a top 10 ranking among OECD countries for unemployment of people with a disability, and ranks a standout first for poverty of people with a disability. What more do you expect Australia can achieve? 🙂

    xasd: Here’s a report on how badly we’re doing on disability. We’re doing very well on a number of other indicators though. AD

  10. SBH

    Ok I’ll bite. But Alan, how can we say countries are coming somewhere when they don’t all compete in the same events? I mean Australia came nowhere in the 100 meters. Whose responsible!! (Long time readers will recognise the rhetorical nature of that question – Of course it’s the carbon tax and Juliar bloody Gillard)

Share this article with a friend

Just fill out the fields below and we'll send your friend a link to this article along with a message from you.

Your details

Your friend's details