Slashdot Mirror
Olympic Medal Prediction Model
bettiwettiwoo writes "Slate reports that PricewaterhouseCooper claims to have devised a model predicting the final medal tally for nations competing in the Olympic Games. GDP is of particular importance in bringing home the bacon, closely followed by population size and and past performance. Other factors can also affect the outcome: hosting the games usually gives a medal boost. With the possible exception of China, the titan nations of the games (US, Russia, China and Germany) are predicted to see a successive drop in their total medal tally in the future (and compared to the Sydney Games, the future starts now). So if you were wondering why the Iraqi soccer team seems on its way to the quarter finals, why Greece takes gold in synchronized diving, or why Michael Phelps has to eat Ian Thorpe's bubbles, don't worry: it's only evolution, baby, and it's all perfectly predictable!"
The variable they seem to have omitted is Propensity of country's sporting bodies to turn blind eye to positive drugs tests."
Thats the primary explaination for the success of the Eastern Europeans in the 60s and 70s, and US Athletics since then.
Athletic Scholarships to universities make as much sense as academic scholarships to sports teams.
How about we forgot about this silly talley and watch the outcome as it unfolds...
Well surely it does. Atheletes are getting better techniques and strategies every games. Otherwise explain how a 4 minute mile was an unacheivable goal 100 years ago and can now be attained by high school students?
Why does Puerto Rico have its own Olympic team?
It's part of the United States, so why? Because it's not a state? No. Washington, DC isn't a state and you don't see it with its own team. This just doesn't make any sense.
A gold medal may require skill, but it needs a whole lot more besides (unfortunately).
For the newer countries entering the competitions, they get better with better facilities and coaching. The US gymnastics got better with the addition of Bella K. The Chinese basketball gets US coaching. International Basketball players get NBA experience and are learning how to trounce the US 'Dream?' team.
Evolution can only be used in this context to explain the improvement of training principles.
Biological evolution would just predict athletes would just get more 'athletier'.
I assumed that the reason Iraq was doing well had to do with the fact that they don't face torture if the return home in defeat. Policy like that has tended to drive the big stars away over the past years.
SPAM
Thorpedo's victory was an upset?
WTF?
Thorpey held the WR, had 9 of the fastest times ever, had not been nbeaten in the distance for 4 years.... add to that Phelps had never gone close to any of Thorpe's times.
Phelps lowered his PB and got third - which, when you look at his performances over the distance is in fact a bloody good result personally for him.
The fact is, it would have been a pretty major upset for Thorpe to lose to Phelps. It was always goignt o be a race between Hoogie and Thorpe, NOT Thorpe and Phelps - it was only moron commentators who were talkign up the clash that begged to differ.
Past performances always said Thorpe verses Hoogie and guess what - that's exactly how it turned out.
Admittedly, the race did live up to hype as an event. It was a damn good one.
Way to stop reading midway through a paragraph, buddy !
The point is backed up in the next line !
The whole quote:
If you're on the poverty line, you don't have a lot of time to invest in sports," said John Hawksworth, head of macroeconomics at PwC. Poor countries like Brazil can excel in soccer, the one truly global sport. But in nations where the horse is still the primary mode of transportation, not many people have the time or leisure to compete in dressage.
Obviously these statistical models aren't trying to pick winners of individual events, but for this race I think the result was pretty much what people expected, despite what Sports Illustrated or Time Magazine might have put on their covers while trying to sell magazines.
What would be realy interesting is to compare how many medals are won per athlete (or team) that participates. Or per person they send to the games, including docters, coaches, trainers and what not.
Also nice would be to compare this with the number of sports they participate in. Crossreference this also with e.g. the amount of people who live in a certain country.
e.g.: The Netherlands will get 21 medals. The US will get 70 medals. Does this mean the US sends more people or that the Dutch are better at sports, if you calculate it per captiva?
Don't fight for your country, if your country does not fight for you.
Statistics isn't about predicting an individual race, no one can do that. It's (partly) about analyzing which dependent variables can accurately predict a response variable, in this case, medal count. While it may be 'common sense', it might be interesting to see for each 100Mil$ GDP, how many more medals on average does a country receive? Questions like that can be answered by the statistical model, not your silly betting questions. And that wasn't even an upset, since the odds-on favorite won the race.
Maybe his win has to do with the fact that the vast majority of aussies live near the ocean.
You North Americans are such bad losers. There are any number of posts here claiming that "other countries" are doing well because they are trained by North American coaches, or because train in the USA.
Now you claim that perhaps the didn't win the swimming because Aussies live near the ocean. Jeeze...
Can't you just accept that sometimes althletes from other countries might be better than the USA ones?
I understand that American media makes a big deal about the total number of medals, because the US has earned a lot of medals, but not many golds. Whereas last I checked Australia and China were dominating in terms of GOLD medals. I think this needs to be more clear.
US beating China?
no way, not now that they had to leave all dopers at home.
BTW, did you notice that as soon as WADA got more active and treated to take more doping tests the "top" american athletes started to disappear??
and not to mention that guy Armstrong, who is "home with his family"....
This might sound good and all, but comon, this just reinforces common sense.
It is not simply reinforcing common sense, it is accurately quantifying the contribution of those factors. That is a awesome feat.
Well I think the athletes futures are important too. Even mild steroids taken frequently can have major impacts upon ones health. But the athletes would be pushed very hard into taking them if they were legal. Not that high doeses of Xenadrine is very safe either, but obvious there are limits. So while I agree the black and white of "cheating" isn't as clear, there has to be limits set, if simply to avoid destroying these athletes in the name of competition.
Tennis (both types), volleyball (especially beach) belong at the Olympics then so does badminton. As much strategy is required for badminton as is for those sports.
Synchronized diving is silly, but I fail to see how it is more so than 'normal' diving.
Woe be on to them, all who rise against poor people, shall perish in a the end. Buju Banton
The Olympics are about skill
No, they're not. If they were, tactical assault, "ultimate fighting", and compettive woodworking would be olympic sports.
The Olympics are, and always have beeen, about "performance." And not in general, but performance in a specific test.
You're absolutely right - the commentators needed to talk up Phelps' attempt for 7 golds - Particularly here in America - which obviously is now over.
That said, Phelps did make it a decent race, as his time was closer to Thorpe's than it was to the 4th place finisher. As you mention, he did set a personal best in his attempt, and there's no shame in that. He also had a real chance at silver (vdH was closer to Phelps than Thorpe).
But ultimately, this was a one-man race from the beginning. And there are some of us Americans whose memories include Sydney and what Thorpe did there. He's not an all around swimmer, but he kills in the free.
-Looking for a job as a materials chemist or multivariat
Statistical estimators being broadcasted without sample variances, t-stats or significance tests.
I mean, would it KILL them to print a standard coefficient table or equation?
Disclaimer: Yes, I teach econometrics.