December 31st, 2004

sleepy

fun with probability

Quote from www.space.com:

Last Thursday, Dec. 23, scientists announced that a space rock named 2004 MN4 had about a 1-in-300 chance of striking Earth on April 13, 2029. On Friday, the risk was upgraded as more observations rolled in. The asteroid was given an unprecedented risk rating of 4 on the Torino Scale, which means it warrants careful monitoring. The odds at various times were put at 1-in-63 and 1-in-45.

As of Monday, the chances of an impact on April 13, 2029 stood at about 1-in-40, or 2.6 percent.

Experts point out that means a 97.4 percent chance the giant boulder will miss, and they stress that the odds are likely to go down to zero, eventually, when more detailed observations of its path are made.

-----

"Likely to go down to zero". Like, say, a 97.4% chance? But unfortunately there's a 2.6% chance that the odds will go to 1 . . .

Saying "1 in 40, but likely to go to zero" is the SAME THING AS SAYING 1 IN 40.

And if you think that, for some reason, there's more than a 1 in 40 chance that the odds will go to 0 . . . then why not factor that into your odds?

(Note: last I heard, the odds were more along the lines of "1 in 500 million", so they did, in fact, effectively go to 0. But my point still stands.)