Lady Crumpet's Armoire


? Oscarzzzzzzzzzzzz | Main | Talking Bush ?

March 04, 2004

It's a Toss-Up. Or is it?

An article in Science News reports that a new analysis suggests that coin tosses are inherently biased - the coin is more likely to land on the face it started with when it's tossed.

There's a discussion about coin tossing in Tom Stoppard's Rosencrantz and Guildenstern are Dead, although I can't remember if it's Rosencrantz or Guildenstern who's consumed about having tossed a coin 99 times and it continually landing on heads (or was it tails?). The point that life is chance, it's all a random toss-up, is slightly turned on its head (or tail) with the discovery that the metaphor is inherently biased. Then again, perhaps Stoppard's Rosencrantz, or Guildenstern, is superhumanly precise and could toss a coin randomly every time.

Damn scientists. ;)

Yours, &c., LC | 12:06 PM | Sundries | Comments (5) | TrackBack (0)

Trackback Pings

TrackBack URL for this entry:


Heads. It's always heads, and it never stops being heads; a brilliant start to a wonderful play.

I played Rosencrantz in college, and he is happily and blissfully unconcerned at each iteration of the coin: it's the major character point for the two of them, to my mind. Ros is overjoyed each time he wins - he always calls heads - and is full of eager hope each time the coin is tossed. Guildenstern, on the other hand, is unconcerned with the momentary bet and the instant toss, and is disturbed at the creepy rebuff of the laws of nature.

Posted by: Linus at March 4, 2004 02:15 PM

i need to see it again, it's been so long. is it on dvd?

Posted by: marco at March 4, 2004 04:42 PM

It is on DVD, or was as I'm not sure it's in print now.

As for the Science News article, I found this interesting: "Their preliminary data suggest that a coin will land the same way it started about 51 percent of the time. It would take about 10,000 tosses before a casual observer would become aware of such a small bias, Diaconis says. 'Maybe that's why society hasn't noticed this before,' he says."

It's things like this that make my eyes roll whenever I see an estimator referred to as "asymptotically unbiased."

Posted by: Scott at March 4, 2004 04:49 PM

It's things like this that make my eyes roll whenever I see an estimator referred to as "asymptotically unbiased."

Which means...?

Posted by: LadyCrumpet at March 5, 2004 09:18 AM

"Asymptotically unbiased" means that with "enough" data the estimate of a particular parameter approaches the actual value of the parameter. What I really want to know is, how much data is enough?

Posted by: Scott at March 5, 2004 09:37 AM