Belgian Euro Coin

Statistics texts talk about coin-flipping trials involving a "fair" coin. It turns out that the Belgian (each country was allowed to put its own national design on one side, kind of like the provincial quarters in Canada or the state quarters of the USA) EuroCoin?s were struck asymmetrically and come up heads about 44% of the time.

• But you can still buy great BelgianBeer with them.

Is there a source for this? you might expect some bias but a 12% swing?

• Do your own homework - Google is your friend:
  • http://www.google.com/search?q=belgian+euro+coin+bias
  • http://www.newscientist.com/article.ns?id=dn1748

Well yes if that is the only source it is almost certainly rubbish (quantitatively), and all reports seem to point to a single Polish experiment that had a sample size of only 250 coin spins. However if I search that myself I have to make an assumption that that is the source quoted above (it does seem fairly likely though).

• There is a way of generating an unbiased outcome from biased coins on RandomNumber.