Finding patterns is easy in any kind of data-rich environment; that's what mediocre gamblers do. The key is in determining whether the patterns represent signal or noise

To be a very, very minor, eighth-tier celebrity, you realize, 'Hey, celebrities are just like us.'

People have different ways of interpreting history.

Not only does political coverage often lose the signal—it frequently accentuates the noise.

When you get into statistical analysis, you don't really expect to achieve fame. Or to become an Internet meme. Or be parodied by 'The Onion' - or be the subject of a cartoon in 'The New Yorker.' I guess I'm kind of an outlier there.

If the state polls are right, then Mr. Obama will win the Electoral College. If you can't acknowledge that after a day when Mr. Obama leads 19 out of 20 swing-state polls, then you should abandon the pretense that your goal is to inform rather than entertain the public.

Every day, three times per second, we produce the equivalent of the amount of data that the Library of Congress has in it's entire print collection, right? But most of it is like cat videos on YouTube or thirteen-year-olds exchanging text messages about the next 'Twilight' movie.