Statisticians Nate Silver and Sam Wang run circles around the pundits
The remarkably accurate predictions of election results by statisticians Nate Silver and Sam Wang has given new credence to number crunching
Tuesday's election result was not just a victory for President Barack Obama, but a validation for the new breed of so-called "big data" number crunchers.
They include New York Times blogger Nate Silver, whose uncanny accuracy in predicting the result has turned him into something of a mathematical rock star.
Silver used his FiveThirty Eight blog to chart the likelihood of victory for Obama or Mitt Romney. His algorithms consistently and clearly pointed to an Obama win - in defiance of conventional wisdom that the race was too close to call - turning him into a hate figure of right-wing pundits, who derided his mathematical "voodoo" as biased.
He was even criticised by one conservative writer for being "thin and effeminate".
Undeterred, on election day he offered a 90.9 per cent probability of an Obama win, predicting 332 as his most likely electoral college tally.
While critics might sneer at correctly predicting a presidential race's outcome, something a coin toss could achieve half the time as Silver himself points out, they have a harder time arguing against the fine print. Silver's model correctly predicted the likeliest presidential outcome in 49 states; it will be correct in all 50 if Obama's lead holds in Florida. That's 332 electoral college votes.
The odds of doing that on a coin toss is less than a one in a thousand trillion. Or, precisely one in 1,125,899,906,842,623.
