What does olympic medal count say about a country? - Page 11

chulbert · 27th August 2012, 07:17 AM

Originally Posted by swright777

If per capita GDP and population are so *********** important, can somebody please explain this?

http://forums.randi.org/imagehosting...a81c7da7e6.jpg
Kenya has the lowest per capita GDP and by far the smallest population of the countries listed. Why is Kenya the only country of the 4 listed that made the top 28 for predicted total medal wins for 2012?

Or this:
http://forums.randi.org/imagehosting...a879404105.jpg

Belarus has 35.6% of Netherlands per capita GDP, and only 56.5% of their population. Why are they both predicted to get the same number of total medals? Hungary has figures much closer to (and slightly better than) Belarus but is only predicted to get 8 medals.

It looks like the previous Olympic results are the only major factor for all countries except Great Britain. They also had the host advantage.

If anyone doesn't get it after this then there's not much more that can be said.

Please try to read this question with as little snark as humanly possible because I'm being really sincere with this:

Do you understand what correlation means? Do you understand what it means to have a correlation between 0 (none) and 1 (perfect)? That there are degrees and strengths of correlation? That finding a handful of exceptions doesn't refute the trend?

It doesn't seem like you do because you're like a dog with a bone when it comes to a handful of data that don't fit the curve. That big countries doing poorly or poor countries doing well, even just a few, somehow completely blows the hypothesis out of the water.

If I made the claim that luxury cars are correlated with income, I don't think there would be much disagreement. It's pretty clear that people who own luxury cars tend to be wealthy. However, there will always be some rich people who drive modest cars, as well as ordinary people who sacrifice in order to have luxury cars.

India and Indonesia are like millionaires that drive Priuses. It's unexpected and they buck the trend but they do not destroy the trend or even relegate it insignificance. Rather than look at the exceptions, look at all the poor, small countries ask whether their overall medal haul generally matches the predictions.

The bottom line is that rich and/or big countries tend to do better than small and/or poor countries. A good mix of both can put you in the top 10 and a bad mix of both is a hole that's very difficult to climb out. It's not always true but it's true enough of the time that it has predictive power, as evidenced by the 91% accurate medal count predictions.

asydhouse · 27th August 2012, 07:30 AM

What does medal haul say about a country?

Absolutely nothing about the character or worth of its people.

I speak as an Englishman, a citizen of the UK, a European, etc

chulbert · 27th August 2012, 07:39 AM

Originally Posted by asydhouse

What does medal haul say about a country?

Absolutely nothing about the character or worth of its people.

I speak as an Englishman, a citizen of the UK, a European, etc

Agreed, strongly. I sincerely hope nobody has ever misconstrued my arguments as veiled jingoism. I'm interested in the subject purely as a mathematical system that can be modeled.

GlennB · 27th August 2012, 08:53 AM

Originally Posted by chulbert

The bottom line is that rich and/or big countries tend to do better than small and/or poor countries. A good mix of both can put you in the top 10 and a bad mix of both is a hole that's very difficult to climb out. It's not always true but it's true enough of the time that it has predictive power, as evidenced by the 91% accurate medal count predictions.

In other words "Things are as they are, and will only change slowly, barring gaining/losing host status". Well, whoopee-doo.

I just compared ranking by total medal count for Beijing v. London and would have got "90% accuracy" just by predicting that nations would hit the same place as last time (allowing +/- 2 places for those lower than 4th in Beijing. But for Japan's rise in London those places would have been +/- 1. But for Japan I could have claimed, oooh, 98% accuracy. Just by copying out the Beijing order.). Shall I change my name to Sherlock?

swright777 · 27th August 2012, 09:07 AM

Originally Posted by chulbert

Please try to read this question with as little snark as humanly possible because I'm being really sincere with this:

Do you understand what correlation means? Do you understand what it means to have a correlation between 0 (none) and 1 (perfect)? That there are degrees and strengths of correlation? That finding a handful of exceptions doesn't refute the trend?

It doesn't seem like you do because you're like a dog with a bone when it comes to a handful of data that don't fit the curve. That big countries doing poorly or poor countries doing well, even just a few, somehow completely blows the hypothesis out of the water.

If I made the claim that luxury cars are correlated with income, I don't think there would be much disagreement. It's pretty clear that people who own luxury cars tend to be wealthy. However, there will always be some rich people who drive modest cars, as well as ordinary people who sacrifice in order to have luxury cars.

India and Indonesia are like millionaires that drive Priuses. It's unexpected and they buck the trend but they do not destroy the trend or even relegate it insignificance. Rather than look at the exceptions, look at all the poor, small countries ask whether their overall medal haul generally matches the predictions.

The bottom line is that rich and/or big countries tend to do better than small and/or poor countries. A good mix of both can put you in the top 10 and a bad mix of both is a hole that's very difficult to climb out. It's not always true but it's true enough of the time that it has predictive power, as evidenced by the 91% accurate medal count predictions.

I'm not saying population and other factors have no effect at all. Just that they are very minor, possibly negligable. And I'm not referring to the skewed results here. What I have a problem with is that they say they use population and GDP as major factors in their predictions, but their predictions do not follow.

Their predictions.

If pop and GDP were major factors in their predictions then I would expect Hungary's expected medal count prediction to be closer to Belarus', and I would expect the Netherlands' prediction to be higher than both of them.

I would also expect the predictions for India, Indonesia and Pakistan to be at least as high as for Kenya.

Since all the predictions are very close to the 2008 results (except GB with their host advantage) for those that they made predictions for, no matter their varying populations or per capita GDP, it would appear that the 2008 results are the only major factor in these predictions. And if they had used the 2008 results as the only factor then their predictions would have been way closer to the actual results.

Also note that they only made predictions for 28 countries, 5 for gold only. Why not the other 176 countries that participated? A mathematical model should have been able to predict all countries. Then we could truly see the (in)accuracy of their process.

chulbert · 27th August 2012, 10:09 AM

Originally Posted by swright777

I'm not saying population and other factors have no effect at all. Just that they are very minor, possibly negligable. And I'm not referring to the skewed results here. What I have a problem with is that they say they use population and GDP as major factors in their predictions, but their predictions do not follow.

Their predictions.

If pop and GDP were major factors in their predictions then I would expect Hungary's expected medal count prediction to be closer to Belarus', and I would expect the Netherlands' prediction to be higher than both of them.

I would also expect the predictions for India, Indonesia and Pakistan to be at least as high as for Kenya.

Since all the predictions are very close to the 2008 results (except GB with their host advantage) for those that they made predictions for, no matter their varying populations or per capita GDP, it would appear that the 2008 results are the only major factor in these predictions. And if they had used the 2008 results as the only factor then their predictions would have been way closer to the actual results.

Also note that they only made predictions for 28 countries, 5 for gold only. Why not the other 176 countries that participated? A mathematical model should have been able to predict all countries. Then we could truly see the (in)accuracy of their process.

Yes, but what caused those past results? What are they based on? The past results are themselves a function of the same factors.

Past results represent a country's tendency to capitalize on their resources (population and GDP) so, yes, it alone has a lot of predictive power.

Nihilianth · 27th August 2012, 10:29 AM

Originally Posted by chulbert

I agree that would be a lot of work. However, that's not at all what the current model is doing so let's please try to stay on track.

Which is exactly why I question the feasibility of such a model.

Quote:

I don't feel comfortable drawing analogies between the overall Olympic games and a single, non-Olympic sport played by children and using that to defend or refute the model. There is no evidence that the same factors play out in the same way in that scenario.

Why does it matter? I admit, luck is a little bit of a stronger factor in children playing a game. But I think 11 and 12 year-olds at the highest level of baseball in the world for that age group makes it not matter.

Some of the same teams from the same areas make it to the World Series year-after-year. Like Tokyo, Japan. Chinese Taipei. Tom's River, New Jersey. Ramstein Air Force Base, Germany.

So obviously "luck" is not all that strong when it comes to the LLWS. If it were, there is no way teams from these very specific cities would be coming here virtually every single year. Or I should say, as often as they do.

chulbert · 27th August 2012, 10:44 AM

Originally Posted by Nihilianth

Why does it matter? I admit, luck is a little bit of a stronger factor in children playing a game. But I think 11 and 12 year-olds at the highest level of baseball in the world for that age group makes it not matter.

It matters because an analogy needs to be applicable to the situation it analogizes. In the LLWS you have a mixture of countries, states, and other organizations, playing a sport that's not in the Olympics, by children at a level far below peak human ability.

I am not at all comfortable drawing any analogy between this and the Olympics. The factors may be similar, some some ways, or they may not. I'm sorry but I just don't see parallels strong enough to warrant comparison.

Nihilianth · 27th August 2012, 03:48 PM

Originally Posted by chulbert

Please try to read this question with as little snark as humanly possible because I'm being really sincere with this:

Do you understand what correlation means? Do you understand what it means to have a correlation between 0 (none) and 1 (perfect)? That there are degrees and strengths of correlation? That finding a handful of exceptions doesn't refute the trend?

It doesn't seem like you do because you're like a dog with a bone when it comes to a handful of data that don't fit the curve. That big countries doing poorly or poor countries doing well, even just a few, somehow completely blows the hypothesis out of the water.

If I made the claim that luxury cars are correlated with income, I don't think there would be much disagreement. It's pretty clear that people who own luxury cars tend to be wealthy. However, there will always be some rich people who drive modest cars, as well as ordinary people who sacrifice in order to have luxury cars.

India and Indonesia are like millionaires that drive Priuses. It's unexpected and they buck the trend but they do not destroy the trend or even relegate it insignificance. Rather than look at the exceptions, look at all the poor, small countries ask whether their overall medal haul generally matches the predictions.

The bottom line is that rich and/or big countries tend to do better than small and/or poor countries. A good mix of both can put you in the top 10 and a bad mix of both is a hole that's very difficult to climb out. It's not always true but it's true enough of the time that it has predictive power, as evidenced by the 91% accurate medal count predictions.

But that wasn't the argument throughout the thread. The argument was trying to disprove "American exceptionalism," and that America was terrible at the Olympics because it's population is so large, and it's wealth so high, and that w didn't win our "fair share" of medals!

The counter-argument has been that, beyond 50 million or so people, population, and GDP doesn't really matter.

My position is that, regardless of population and income level, it all comes down to three things: culture, desire to play, and of course location. Only GDP would have any sort of significant effect. NOT population so much.

The nation of Norway proves my hypothesis about the insignificance of population and the significance of culture and location. They are great at winter sports. The nation of Kenya downplays the significance of GDP. They are great runners. The nations of the USA proves the effect of Olympic rules. We are barred from sending a proportional number of athletes to our population. The game of basketball disproves the theory that sending more athletes would decrease your chances of medaling (word?) in a particular team sport. We have a dozen or more NBA players that could very easily replace Kobe. The games of American football and baseball proves the insignificance of Olympic medal haul as a sign of overall national athleticism. The Olympics do not even feature the two most popular American sports.

Nihilianth · 27th August 2012, 04:04 PM

Originally Posted by chulbert

It matters because an analogy needs to be applicable to the situation it analogizes. In the LLWS you have a mixture of countries, states, and other organizations, playing a sport that's not in the Olympics, by children at a level far below peak human ability.

I am not at all comfortable drawing any analogy between this and the Olympics. The factors may be similar, some some ways, or they may not. I'm sorry but I just don't see parallels strong enough to warrant comparison.

What does "Peak human ability" matter? You do realize that "peak human ability" is on a bell curve, right? Most of those 11 and 12 year olds are currently at their peak. Even most of the LLWS participants. Not to mention, that is a slight move of the goal posts.

Looking at 12 year olds for a world-wide popular sport such as baseball, is a great way to analyze the GRASSROOTS of talent. After all, some of them will become the future's baseball stars. And virtually all of the future baseball stars are currently 12. Of course, there are outliers of people who have never touched a baseball in their lives until going to college, and becoming great ballplayers. But that is EXTREMELY rare. So rare, in fact, as to be negligible.

But yes. We are talking about anti-American exceptionalism, and using the Olympics as the gold-standard for proving this. Even though it has been shown that, first, the Olympics naturally make it impossible for any one country to get past a certain number of medals. Second, that the Olympics don't even feature many nations' most popular sports. (Shocking that they got rid of baseball, as it is the most popular sport internationally, obviously second only to soccer.) And third, there are other, non-sports-related ways to prove "American exceptionalism" does not exist.

i have broadened the horizone to include other sports not in the Olympics, simply because talking about the Olympics is not at all productive. Little League Baseball is every bit as much a sport as the Fifa World Cup. And Little League Baseball is the single largest sports organization in the world, encompassing over 100 nations, and 16 Regions. And yet, the same Little Leagues make it far more often than the vast majority of Little Leagues. Which also proves the "desire to play," and the "cultural effect." It also proves the power of coaching. Population and even GDP has very, very little to do with it. (You would think that with the large number of Little Leagues in cities like NYC, and Philly would have a much greater chance of fielding a WS-level team. But they never do.)

27th August 2012, 08:53 AM	#404
GlennB Jellied eel and offal fancier Join Date: Sep 2006 Location: Arcadia Posts: 8,954	Originally Posted by chulbert The bottom line is that rich and/or big countries tend to do better than small and/or poor countries. A good mix of both can put you in the top 10 and a bad mix of both is a hole that's very difficult to climb out. It's not always true but it's true enough of the time that it has predictive power, as evidenced by the 91% accurate medal count predictions. In other words "Things are as they are, and will only change slowly, barring gaining/losing host status". Well, whoopee-doo. I just compared ranking by total medal count for Beijing v. London and would have got "90% accuracy" just by predicting that nations would hit the same place as last time (allowing +/- 2 places for those lower than 4th in Beijing. But for Japan's rise in London those places would have been +/- 1. But for Japan I could have claimed, oooh, 98% accuracy. Just by copying out the Beijing order.). Shall I change my name to Sherlock?
	Last edited by GlennB; 27th August 2012 at 08:56 AM.

27th August 2012, 10:09 AM	#406
chulbert Muse Join Date: May 2002 Location: Rochester, NY Posts: 725	Originally Posted by swright777 I'm not saying population and other factors have no effect at all. Just that they are very minor, possibly negligable. And I'm not referring to the skewed results here. What I have a problem with is that they say they use population and GDP as major factors in their predictions, but their predictions do not follow. Their predictions. If pop and GDP were major factors in their predictions then I would expect Hungary's expected medal count prediction to be closer to Belarus', and I would expect the Netherlands' prediction to be higher than both of them. I would also expect the predictions for India, Indonesia and Pakistan to be at least as high as for Kenya. Since all the predictions are very close to the 2008 results (except GB with their host advantage) for those that they made predictions for, no matter their varying populations or per capita GDP, it would appear that the 2008 results are the only major factor in these predictions. And if they had used the 2008 results as the only factor then their predictions would have been way closer to the actual results. Also note that they only made predictions for 28 countries, 5 for gold only. Why not the other 176 countries that participated? A mathematical model should have been able to predict all countries. Then we could truly see the (in)accuracy of their process. Yes, but what caused those past results? What are they based on? The past results are themselves a function of the same factors. Past results represent a country's tendency to capitalize on their resources (population and GDP) so, yes, it alone has a lot of predictive power.
	Last edited by chulbert; 27th August 2012 at 10:14 AM.

27th August 2012, 10:29 AM	#407
Nihilianth Master Poster Join Date: Feb 2011 Posts: 2,224	Originally Posted by chulbert I agree that would be a lot of work. However, that's not at all what the current model is doing so let's please try to stay on track. Which is exactly why I question the feasibility of such a model. Quote: I don't feel comfortable drawing analogies between the overall Olympic games and a single, non-Olympic sport played by children and using that to defend or refute the model. There is no evidence that the same factors play out in the same way in that scenario. Why does it matter? I admit, luck is a little bit of a stronger factor in children playing a game. But I think 11 and 12 year-olds at the highest level of baseball in the world for that age group makes it not matter. Some of the same teams from the same areas make it to the World Series year-after-year. Like Tokyo, Japan. Chinese Taipei. Tom's River, New Jersey. Ramstein Air Force Base, Germany. So obviously "luck" is not all that strong when it comes to the LLWS. If it were, there is no way teams from these very specific cities would be coming here virtually every single year. Or I should say, as often as they do.
	Last edited by Nihilianth; 27th August 2012 at 10:30 AM.

27th August 2012, 03:48 PM	#409
Nihilianth Master Poster Join Date: Feb 2011 Posts: 2,224	Originally Posted by chulbert Please try to read this question with as little snark as humanly possible because I'm being really sincere with this: Do you understand what correlation means? Do you understand what it means to have a correlation between 0 (none) and 1 (perfect)? That there are degrees and strengths of correlation? That finding a handful of exceptions doesn't refute the trend? It doesn't seem like you do because you're like a dog with a bone when it comes to a handful of data that don't fit the curve. That big countries doing poorly or poor countries doing well, even just a few, somehow completely blows the hypothesis out of the water. If I made the claim that luxury cars are correlated with income, I don't think there would be much disagreement. It's pretty clear that people who own luxury cars tend to be wealthy. However, there will always be some rich people who drive modest cars, as well as ordinary people who sacrifice in order to have luxury cars. India and Indonesia are like millionaires that drive Priuses. It's unexpected and they buck the trend but they do not destroy the trend or even relegate it insignificance. Rather than look at the exceptions, look at all the poor, small countries ask whether their overall medal haul generally matches the predictions. The bottom line is that rich and/or big countries tend to do better than small and/or poor countries. A good mix of both can put you in the top 10 and a bad mix of both is a hole that's very difficult to climb out. It's not always true but it's true enough of the time that it has predictive power, as evidenced by the 91% accurate medal count predictions. But that wasn't the argument throughout the thread. The argument was trying to disprove "American exceptionalism," and that America was terrible at the Olympics because it's population is so large, and it's wealth so high, and that w didn't win our "fair share" of medals! The counter-argument has been that, beyond 50 million or so people, population, and GDP doesn't really matter. My position is that, regardless of population and income level, it all comes down to three things: culture, desire to play, and of course location. Only GDP would have any sort of significant effect. NOT population so much. The nation of Norway proves my hypothesis about the insignificance of population and the significance of culture and location. They are great at winter sports. The nation of Kenya downplays the significance of GDP. They are great runners. The nations of the USA proves the effect of Olympic rules. We are barred from sending a proportional number of athletes to our population. The game of basketball disproves the theory that sending more athletes would decrease your chances of medaling (word?) in a particular team sport. We have a dozen or more NBA players that could very easily replace Kobe. The games of American football and baseball proves the insignificance of Olympic medal haul as a sign of overall national athleticism. The Olympics do not even feature the two most popular American sports.
	Last edited by Nihilianth; 27th August 2012 at 03:49 PM.

27th August 2012, 04:04 PM	#410
Nihilianth Master Poster Join Date: Feb 2011 Posts: 2,224	Originally Posted by chulbert It matters because an analogy needs to be applicable to the situation it analogizes. In the LLWS you have a mixture of countries, states, and other organizations, playing a sport that's not in the Olympics, by children at a level far below peak human ability. I am not at all comfortable drawing any analogy between this and the Olympics. The factors may be similar, some some ways, or they may not. I'm sorry but I just don't see parallels strong enough to warrant comparison. What does "Peak human ability" matter? You do realize that "peak human ability" is on a bell curve, right? Most of those 11 and 12 year olds are currently at their peak. Even most of the LLWS participants. Not to mention, that is a slight move of the goal posts. Looking at 12 year olds for a world-wide popular sport such as baseball, is a great way to analyze the GRASSROOTS of talent. After all, some of them will become the future's baseball stars. And virtually all of the future baseball stars are currently 12. Of course, there are outliers of people who have never touched a baseball in their lives until going to college, and becoming great ballplayers. But that is EXTREMELY rare. So rare, in fact, as to be negligible. But yes. We are talking about anti-American exceptionalism, and using the Olympics as the gold-standard for proving this. Even though it has been shown that, first, the Olympics naturally make it impossible for any one country to get past a certain number of medals. Second, that the Olympics don't even feature many nations' most popular sports. (Shocking that they got rid of baseball, as it is the most popular sport internationally, obviously second only to soccer.) And third, there are other, non-sports-related ways to prove "American exceptionalism" does not exist. i have broadened the horizone to include other sports not in the Olympics, simply because talking about the Olympics is not at all productive. Little League Baseball is every bit as much a sport as the Fifa World Cup. And Little League Baseball is the single largest sports organization in the world, encompassing over 100 nations, and 16 Regions. And yet, the same Little Leagues make it far more often than the vast majority of Little Leagues. Which also proves the "desire to play," and the "cultural effect." It also proves the power of coaching. Population and even GDP has very, very little to do with it. (You would think that with the large number of Little Leagues in cities like NYC, and Philly would have a much greater chance of fielding a WS-level team. But they never do.)
	Last edited by Nihilianth; 27th August 2012 at 04:09 PM.

27th August 2012, 07:17 AM	#401
chulbert Muse Join Date: May 2002 Location: Rochester, NY Posts: 725	Originally Posted by swright777 If per capita GDP and population are so *********** important, can somebody please explain this? http://forums.randi.org/imagehosting...a81c7da7e6.jpg Kenya has the lowest per capita GDP and by far the smallest population of the countries listed. Why is Kenya the only country of the 4 listed that made the top 28 for predicted total medal wins for 2012? Or this: http://forums.randi.org/imagehosting...a879404105.jpg Belarus has 35.6% of Netherlands per capita GDP, and only 56.5% of their population. Why are they both predicted to get the same number of total medals? Hungary has figures much closer to (and slightly better than) Belarus but is only predicted to get 8 medals. It looks like the previous Olympic results are the only major factor for all countries except Great Britain. They also had the host advantage. If anyone doesn't get it after this then there's not much more that can be said. Please try to read this question with as little snark as humanly possible because I'm being really sincere with this: Do you understand what correlation means? Do you understand what it means to have a correlation between 0 (none) and 1 (perfect)? That there are degrees and strengths of correlation? That finding a handful of exceptions doesn't refute the trend? It doesn't seem like you do because you're like a dog with a bone when it comes to a handful of data that don't fit the curve. That big countries doing poorly or poor countries doing well, even just a few, somehow completely blows the hypothesis out of the water. If I made the claim that luxury cars are correlated with income, I don't think there would be much disagreement. It's pretty clear that people who own luxury cars tend to be wealthy. However, there will always be some rich people who drive modest cars, as well as ordinary people who sacrifice in order to have luxury cars. India and Indonesia are like millionaires that drive Priuses. It's unexpected and they buck the trend but they do not destroy the trend or even relegate it insignificance. Rather than look at the exceptions, look at all the poor, small countries ask whether their overall medal haul generally matches the predictions. The bottom line is that rich and/or big countries tend to do better than small and/or poor countries. A good mix of both can put you in the top 10 and a bad mix of both is a hole that's very difficult to climb out. It's not always true but it's true enough of the time that it has predictive power, as evidenced by the 91% accurate medal count predictions.

27th August 2012, 07:30 AM	#402
asydhouse Muse Join Date: Feb 2012 Location: Swansea in the UK Posts: 533	What does medal haul say about a country? Absolutely nothing about the character or worth of its people. I speak as an Englishman, a citizen of the UK, a European, etc

27th August 2012, 07:39 AM	#403
chulbert Muse Join Date: May 2002 Location: Rochester, NY Posts: 725	Originally Posted by asydhouse What does medal haul say about a country? Absolutely nothing about the character or worth of its people. I speak as an Englishman, a citizen of the UK, a European, etc Agreed, strongly. I sincerely hope nobody has ever misconstrued my arguments as veiled jingoism. I'm interested in the subject purely as a mathematical system that can be modeled.

27th August 2012, 09:07 AM	#405
swright777 Muse Join Date: Feb 2012 Location: Pennsylvania, USA Posts: 576	Originally Posted by chulbert Please try to read this question with as little snark as humanly possible because I'm being really sincere with this: Do you understand what correlation means? Do you understand what it means to have a correlation between 0 (none) and 1 (perfect)? That there are degrees and strengths of correlation? That finding a handful of exceptions doesn't refute the trend? It doesn't seem like you do because you're like a dog with a bone when it comes to a handful of data that don't fit the curve. That big countries doing poorly or poor countries doing well, even just a few, somehow completely blows the hypothesis out of the water. If I made the claim that luxury cars are correlated with income, I don't think there would be much disagreement. It's pretty clear that people who own luxury cars tend to be wealthy. However, there will always be some rich people who drive modest cars, as well as ordinary people who sacrifice in order to have luxury cars. India and Indonesia are like millionaires that drive Priuses. It's unexpected and they buck the trend but they do not destroy the trend or even relegate it insignificance. Rather than look at the exceptions, look at all the poor, small countries ask whether their overall medal haul generally matches the predictions. The bottom line is that rich and/or big countries tend to do better than small and/or poor countries. A good mix of both can put you in the top 10 and a bad mix of both is a hole that's very difficult to climb out. It's not always true but it's true enough of the time that it has predictive power, as evidenced by the 91% accurate medal count predictions. I'm not saying population and other factors have no effect at all. Just that they are very minor, possibly negligable. And I'm not referring to the skewed results here. What I have a problem with is that they say they use population and GDP as major factors in their predictions, but their predictions do not follow. Their predictions. If pop and GDP were major factors in their predictions then I would expect Hungary's expected medal count prediction to be closer to Belarus', and I would expect the Netherlands' prediction to be higher than both of them. I would also expect the predictions for India, Indonesia and Pakistan to be at least as high as for Kenya. Since all the predictions are very close to the 2008 results (except GB with their host advantage) for those that they made predictions for, no matter their varying populations or per capita GDP, it would appear that the 2008 results are the only major factor in these predictions. And if they had used the 2008 results as the only factor then their predictions would have been way closer to the actual results. Also note that they only made predictions for 28 countries, 5 for gold only. Why not the other 176 countries that participated? A mathematical model should have been able to predict all countries. Then we could truly see the (in)accuracy of their process.

27th August 2012, 10:44 AM	#408
chulbert Muse Join Date: May 2002 Location: Rochester, NY Posts: 725	Originally Posted by Nihilianth Why does it matter? I admit, luck is a little bit of a stronger factor in children playing a game. But I think 11 and 12 year-olds at the highest level of baseball in the world for that age group makes it not matter. It matters because an analogy needs to be applicable to the situation it analogizes. In the LLWS you have a mixture of countries, states, and other organizations, playing a sport that's not in the Olympics, by children at a level far below peak human ability. I am not at all comfortable drawing any analogy between this and the Olympics. The factors may be similar, some some ways, or they may not. I'm sorry but I just don't see parallels strong enough to warrant comparison.