I’m surprised at how bad the predictions were for the recent Olympics – perhaps the home court advantage was stronger than he expected it to be? Or perhaps he needs a different model for summer versus winter olympics?

I was an econ major in college, and I love taking looks at stats like these, since I have a background in statistics, and now with software development, I am able to manipulate datasets

Or you’re right – perhaps we just can’t account for miracles in medal counts, and there will always be some outlier cases that beat the odds =P

Thank you for for using my photograph in this post!

Please attribute the photograph to Duncan Rawlinson and link to me @ http://www.TheLastMinuteBlog.com

Thank you.

When I first started looking into investing in stocks in the mid 90s, what I heard from everyone was “The market used to average 10% a year in the long run, but that was in the old economy. The internet changed everything. In the new economy, it averages 15% a year in the long run.”

Of course, it didn’t last. I wonder what economic models predicted that it would.

@ Ian, I don’t know if I’ve seen a formula that attempted to account for choice. I just started reading your book Choice – The Meaning of Life, so I’m interested in seeing what you have to say about how we make choices.

I hadn’t heard of circular breathing, and I would have thought it was impossible. I sure can’t do it, but I guess it’s supposed to be hard.

There are too many variables for such a prediction and by his own admittance China, who wins a lot of medals, throws it off. (So we just ignore China.) You can’t account for passion, drive, spirit, etc, especially in sports that don’t lend themselves to the most stacked teams winning. The U.S. hockey team is a great example. No way that team wins silver, let alone almost gold, on paper.

People are always disproving not only formulas but science itself. Formulas fail to measure one thing: Choice. They make it look like you have no choice, as you note when you say “just hand out the medals” without competing. (Incidentally, in their previous two Olympics as host, Canada didn’t win one gold, which really throws a wrench into the professor’s formula.) A great example is wind musicians. Science claimed for ages that you can’t exhale or inhale simultaneously. We used to argue with our teachers about it in high school because we had trumpet players in band who could “circular breath,” inhale while exhaling, holding notes for several minutes. I had a biology professor who called me a flat out liar! Then Kenny G came along years later and held a note for over a day… pretty sure he had to breathe while doing it! These musicians made a choice to go against the scientific rules of the body and got their body to do what they wanted. Remarkable! And something that won’t show up in any formula.

The interesting thing with economics is that I bet people with decent experience in whatever field the economist is modeling could be as accurate with off-the-cuff predictions.

For example, with the olympics, I’m fairly confident I could guess and get an accuracy rate of 80% and up. My method would take 15-20 minutes total (guessing for each country based on prior olympic medal totals) and this economists method took how long to develop? Days? Weeks? Sometimes I wonder if we try to be too precise with a lot of things and make them more expensive or labor intensive than necessary.

Another example would be the housing bubble. When people making $100k a year were having trouble buying a house in half the country it seems rather obvious something was wrong. It’s not like the U.S. population doubled in 5 years and space was at a premium. How much more info did these economists who failed to predict this event need?

But this claim was based on its return over the 20 year period ending in 1994, or something like that. They just looked at that range of the past data, and found a formula that worked over that time. But it didn’t work before that period, and there wasn’t any data to see if it would work going forward (it didn’t).

It is also kind of like in physics, which suggests you could predict everything if you knew every bit of information about every particle in the universe. However, just observing the particle changes its properties.

Just knowing what the models look for causes people to artificially change those variables.