Olympic News
August 24 update:
Congratulations to Beijing on a successful Olympics! Our predictions
ended up 93% on-target (92% for gold medals). China won the gold medal
race (as predicted), the US won the most medals overall (as predicted), and
both nations surpassed our predictions a scant 24 hours before the Games
concluded.
So who won? I would
argue that neither China nor the US 'won' the Games, as both were expected
to perform well. After all, resources matter. The true 'winning
nations' should be the ones that did amazing things with the resources given
to them. That is, who outperformed expectations by the widest margin?
The answer, including all of our predictions and
post-Games analysis of where they failed, are linked
here.
There are even some cool graphs to entertain you.
If you really cannot wait
10 seconds for the download, here's my answer:
-
Kenya, Canada,
Denmark, Belgium and Zimbabwe all won gold medals but were not predicted
to do so (incidentally, Kenya won 5 and Canada won 3). That makes
them the ultimate 'winners' in my book.
-
-
Zimbabwe, Iceland, Sudan,
Togo and Venezuela all won medals but were not predicted to do so (Zimbabwe
won 4). That makes them all 'winners' in my book as well.
-
-
Among larger
medal-winning nations, Australia and Great Britain both beat the total medal
predictions by over 68 percent. In the gold medal race, Australia won
12 more gold medals than predicted, while Great Britain won 16 more than
expected. Kudos to those squads!
If you want to talk more
about the results, feel free to
email me.
Now that the Games are over, I suddenly have some free time on my hands.
August 17 update: Predictions for the Beijing 2008 Games are
right on target! Halfway through the medal awards, our predictions
have an 88% correlation with actual medal standings (87% for gold medals
alone).
These are predictions based on a model
created with Ayfer Ali
before the 2000 Olympic Games in Sydney (see the link to the paper below).
Click
here
to access the 2008 predictions.
We will post again here right after the Beijing Games with a synopsis of
where we were close, where we weren't close, and what that all means.
July 1, 2008: Big
news! Our model predicts that we will hear the Chinese national anthem more
than any other nation's anthem, with China winning 44 gold medals this
year. That's a huge
prediction when you consider three things:
1) the next nation will be
the U.S. with 33 (equal to the U.S. gold medal haul in Athens in 2004) but
2)
China has only won 112 gold medals at all previous Summer Olympics venues
combined, while
3) the post-War record for most gold medals at a non-boycotted Summer
Games is 50 by the former Soviet Union in 1972. If our predictions
hold, China would tie for the 'third most golden nation in Olympic history',
and would tie the U.S. for the most golden host nation ever (as the U.S. won
44 gold medals in Atlanta in 1996).
The
U.S. will still win the most medals overall (103, equal to the Athens
performance), Russia will win 95 (3 more than in Athens), and China will win
89 medals overall.
Previous Olympics and predictions: We
decided to continue our work, since our
predictions
for Torino 2006 were accurate (again). In Torino, we
correctly predicted 13 of the 14 top medal-winning nations, including a
forecast of Germany at the top of the standings. Our forecast was only
1 medal away
from the final German total. Overall,
we found a 0.93 correlation
between forecasts and actual medals across all participant nations (0.89
correlation for gold medals
alone).
For
the U.S. in particular, we predicted 22 U.S. medals (the actual count was
25), substantially closer than any other forecast of which we are aware. We
forecast 8 U.S. gold medals, compared to 9 actually won.
Our Athens
2004 Summer Games predictions were
surprisingly accurate as well. For
example, we predicted 103 U.S. medals, 37 of them gold. The American
team won precisely 103 medals, 35 of them gold. We predicted 94
Russian medals, and the Russian contingent won 92. Details are
available here.
Using only income per capita, population,
climate, and political structure (along with nation-specific effects), we
have explained participation and
medal counts at the post-War Olympic Games. We have published
out-of-sample predictions before each of the last four venues:
Medal
counts
Torino
2006 Winter Games had a 0.93 correlation
with our predicted results
Athens 2004 Summer
Games had a 0.94 correlation with our predicted results
Salt Lake 2002 Winter
Games had a 0.94 correlation with our predicted results
Sydney 2000 Summer Games
had a 0.95 correlation with our predicted results
Gold
medal counts
Torino
2006 Winter Games had a 0.89 correlation
with our predicted results
Athens 2004 Summer Games
had a 0.86 correlation with our predicted results
Salt Lake 2002 Winter
Games had a 0.85 correlation with our predicted results
Sydney 2000 Summer Games
had a 0.84 correlation with our predicted results
The conclusion?
There is a measurable, continuing
advantage to certain nations in the Olympic Games. We should be aware of that fact when we
compare outcomes across nations, and when we set our own national goals for medal counts.
So
who actually "won" the medal race in Torino 2006? Canada and
China both outperformed the forecast by more than fifty
percent.
Download our
data here (citing the paper as linked above).
A
history of selected press coverage is available
here.
|
