*Chart of the Day - 11.27.2008

good one kevin. you seem to getting your mojo back.

By far (3) is the scarriest. It is easy to imagine that participants would expect some combination of cheating/bad-luck would cause them to get a lower than expected payoff.

Coins obey newtonian physics so the answer really depends on the experiment. 3.9 sounds about right if you are spinning pennies. On the other hand if I stand them all on their side and bump the table from a few feet away I'm pretty sure I could get all heads. Catching a flipped coin takes away a significant bias.

I remember getting detention from our 4th grade math teacher for confidently telling her that tails was more frequent when tossing pennies on the ground. She kept us quite a while but gave up when we were at about 600 tails and 400 heads.

And sure enough the last sentence of the paper is:

The elicited pessimism could be interpreted as defensive pessimism, an
anticipatory strategy that involves setting defensively low expectations prior to entering
a situation so as to defend against loss of self esteem in the event of failure.

So a different experiment should have been set up -- one in which points (or euros) were lost as the actual score deviated from the stated expectation.

What about those 25 who think they are going to lose every time? That's not exactly encouraging. Dare to hope, people...

And there are those who probably suspected it was a trick question.

Euro coins have no heads.

If they'd been business news pundits, they'd have sold tickets to their upcoming seminar on how to win 11 times out of 10 - with no money down!!!

HAhaha! Perfect!

I think this is a bad experiment. They're not measuring what they think they're measuring.

Say a stranger comes up to you and tells you he will give you a good deal of money every time a coin comes up heads. The social situation is overwhelming your rational estimation. You want to get-along, tell the person they won't have to give you too much money. You want to show a little humility, a little self-deprecation about your luck. It's telling that women, who are more socially aware than men, would be more prone to this effect.

They should recreate this experiment and conduct it through a computer terminal. I would predict that the effect would disappear entirely.

Kevin Drum: The exception, of course, is those ten respondents who think they'll win every time. Here in America, we call those people "investment bankers."

Enough demagoguery. Investment bankers are the best and brightest, educated at our country's finest universities. They know that the expected value is 5.0. Using sophisticated mathematical modeling, they will improve the odds of this primitive game by dividing the tosses into tranches and selling them to other sophisticated investors. Expected returns will be increased to 500 euros by appropriate leverage. To hedge against the unlikely they then become a counterparty to as many as ten other counterparties for each toss (nine of whom aren't even otherwise involved in the transaction).

And I haven't even gotten to the sophisticated part! The key step is to then convince the US government that what is technically known as "male bovine manure" is equivalent to "unforeseeable consequences" (this step explains why a narrow technical education is inappropriate, and must be rounded out by subjects such as the theater arts). Lastly, it's essential that the audience be convinced that "panhandling" and "welfare" are not appropriate synonyms for "saving the economy".

Ok, given the supine US government and the ready assistance of the MSM and certain serious bloggers, that last step is surprisingly easy. Nevertheless, taken as a whole, it's quite a trick.

#3 is scary because many economists assume that economic decisions are rational and informed. It should be common knowledge that they are not, and if this were common knowledge then the results of this simple study would not be worth publishing.

As a matter of fact, it is well known to some scientists (not economists, who are not scientists anyway) that the degree of rationality of decisions is highly sensitive to social context.

some do, like the euros from monarchies like spain or the netherlands

Very true -- but still, many do not, so if you take "heads" literally you might be skeptical about which coin they would choose in this particular study.

By the way, the heads on the 1-euro coins are not all monarchs -- they include Mozart, the Pope, and various national heroes.

Richard Feynman (the Nobel-winning physicist) once spent some time in Vegas, hung out at the craps table, and met a guy who was a professional gambler. Feynman asked the man how he could make money gambling, since the odds were in favor of the house. He explained:

"I'm standing around a table, when some guy says, 'It's comin' out nine! It's gotta be a nine!' The guy's excited; he thinks it's going to be a nine, and he wants to bet. Now I know the odds for all the numbers inside out, so I say to him, 'I'll bet you four to three it's not a nine,' and I win in the long run."