Thursday, September 13, 2012

Olympics 2012 - Another Perspective II

Courtesy: Geckoam
(Not) A Game of Money
Continuation of Olympics 2012 - Another Perspective I

Does rich nations play more sports? And do they win more than their poor counterparts? Let's analyze the data for answers to these questions.

GDP Per Participant:  If I have more money at disposal, I would spend more time and money on sports and hence, would rank higher in terms of number of participants and medals. A poor country which secured one medal would have done it against innumerable odds compared to a wealthy nation and should actually rank higher. Taking the wealth of the citizens into account, we can rank nations based on how many athletes it produced per dollar of their gross domestic product (GDP).

For this purpose, I used GDP at Purchasing Power Parity Per Capita instead of nominal GDP. The reason being the GDP in itself would reflect the national wealth, not general prosperity of its citizen. A per capita GDP (PPP), though not equivalent to per capita income, does reflect the wealth of citizen while adjusting for varying cost of living across geographies. Now, getting to the ranks.

Best GDP to Athlete Ratio: China - 22 USD per capita per participant.
Worst GDP to Athlete Ratio: Liechtenstein - 47,033 USD per capita per participant.

So, China produces the maximum number of participants per dollar whereas Liechtenstein, the richest participating nation, produced the least. And India? Well, India did a rather good job here, standing at 7th position, spending 45 USD per capita per participant!

Correlation Coefficient - GDP and No. of Participants: 0.2626

Does rich nations play more? Yes, they do, but poor countries do as much too. The correlation coefficient which, even though positive, is just a quarter and signifies that the correlation is not that strong. Even though sports requires money, we can play with little money as well. Next time you are told that the lack of fund is sololy responsible for the pathetic state of Indian hockey, tell them about the correlation and I bet they would still continue their rant because it is easy to understand what lack of fund means but not correlation coefficient.

GDP Per Medal: Whereas GDP Per Participant tells us about how efficienctly the nation uses its resources to promote sports, GDP Per Medal tells us about the financial efficiency in winning games. Again, just like Population Per Medal, we are comparing only the 85 nations that won any medals.

Best GDP to Medal ratio: China - 95 USD per capita per participant.
Worst GDP to Medal ratio: Qatar - 51,350 USD per capita per participant.

China did it again, winning maximum number of medals per dollar of GDP. The filthy rich Qatar did worst on GDP to medal ratio amongst the 85 countries that won any medal.

Correlation Coefficient - GDP and Medals: 0.1938

Does rich nations win more than poor folks? As it is evident from the correlation coefficient, there is little correlation between individual wealth and games won. In other words, wealthy first world countries are very slightly more likely to win games than underdeveloped countries.
Even though money is a crucial factor in playing sports, there may be other factors at play which compensate for lack of money. What are those factors? The numbers won't tell that. And what is your guess?

As the numbers talk, it is evident that there is no clear relation between performance in sports and wealth. The potential of a middle-class kid next door transforming into a sport star is only slightly less than that of wealthy folks. At least, this is what the Olympics data suggests. To know the actual probability of it and the correlation between a wealthy family and sports performance, we would need a different set of data.

Participant Per Medal: A nation can field a large number of participants. Though it does improve the probability of winning more medals, the quality of participants have a rather major role to play. In my opinion, a nation which fileded ten participants and won 5 medals should rank higher than a nation which fielded hundred participants but won only 30 medals. Here is a comparison of nations based on how many quality participant they produced worth their salt.

Best Participant to Medal ratio: Botswana, China, Iran, Jamaica, Kenya - 4 participants per medal.
Worst Participant to Medal ratio: Portugal - 77 participants per medal.

India had a participant to medal ratio of 14 and ranked 47 out of 85 medal-winning nations.

In summary, China seems to have done well whether it is about utilizing money or about producing quality participants.

Correlation Coefficient - Participants and Medals: 0.897774

Now, this is an obvious conclusion. The more the number of participants, the more the medals. However, the point to point finger would be that there is no perfect correlation between the number of participants and medals and in fact, it is a tad less than 1. What it implies is nations with higher number of participants won less medals per participant when compared to the averages.

For a detailed look on the numbers, check out the spreadsheet below.
Save on Delicious

No comments:

Post a Comment

Creative constructive criticism is accepted and expected.