THE BIG READ: Silver's statistical gold - Times LIVE
Sat Apr 29 15:24:34 SAST 2017

THE BIG READ: Silver's statistical gold

Andrew Hough | 2012-11-09 00:06:56.0
The GE Building displays early results of the US presidential election in Rockefeller Centre, New York City. Many pundits and polsters' predictions were way off Picture: MIKE COPPOLA/GALLO IMAGES

Nate Silver, who writes the FiveThirtyEight blog for the New York Times, basked in worldwide adulation after his polling correctly forecast the outcome of the US presidential election in 49 states.

The statistician, who tracked the president's statistical odds, will have correctly foreseen the result in every US state.

On election day, the 34 year-old, who had become a target of conservatives, also offered a 90.9% probability of an Obama win in the US election.

He forecast that Obama would win 332 electoral college votes compared to 206 for Mr Romneya.

In the dying days of the campaign Silver came under fire from Republicans for arguing that President Obama's chance of winning was rapidly increasing.

His blog, which is named after the number of electoral college votes up for grabs, was pilloried by conservative commentators during the campaign.

But political commentators this week universally hailed his polling forecasts for their accuracy, saying that he was the other main winner in an extraordinary US presidential campaign. He is said to be an Obama supporter.

"You know who won the election tonight? - Nate Silver," said Rachel Maddow, the television host, during MSNBC's election coverage.

In an opinion column on the website Mashable, Chris Taylor wrote: "Here is the absolute, undoubted winner of this election: Nate Silver and big data. What does this victory mean? That mathematical models can no longer be derided by 'gut-feeling' pundits.

"Silver's contention - that TV pundits are generally no more accurate than a coin toss - must now be given wider credence."

The predictions by Silver and other analysts - who for the technically literate use techniques known as quantitative forecasting or Bayesian analysis, named after mathematician Thomas Bayes - drew wide audiences during the campaign and prompted many people to place bets on accuracy.

Former Republican congressman and cable news host Joe Scarborough called Silver an "ideologue" on television and bet Silver $1000 on the outcome of the race.

"Anybody who thinks that this race is anything but a toss-up right now is such an ideologue, they should be kept away from typewriters, computers, laptops and microphones for the next 10 days, because they're jokes," he said.

Silver tweeted at him last Thursday, stating: "If you think it's a toss-up, let's bet. If Obama wins, you donate $1000 to the American Red Cross. If Romney wins, I do. Deal?"

Silver, also a former economist, first made his mark with a computer model that predicts baseball player performance.

After graduating from the University of Chicago in 2000, Silver worked as an economics consultant at an accounting firm before creating a model for predicting baseball players' performance. He sold it to statistics firm Baseball Prospectus for an undisclosed large amount.

He turned to politics in 2007 with a model he devised that relies on state-by-state polling.

Silver incorporates the averages of several polls and weights them based on factors such as the past accuracy of the polling firm, the number of likely voters on election day and the composition of each state's electorate. He then runs multiple simulations of the results, which results in his probability forecast.

The end result often mirrors other aggregate data that are available.

Real Clear Politics and, for instance, also showed that Obama held an advantage in all of the swing states except North Carolina. But Silver's probability simulations, as well as his status as, essentially, a one-man shop, has helped burnish his image and reputation, especially in light of the performance of traditional polling firms.

Rasmussen Reports, for instance, was wrong on six of the nine swing-state polls and showed Romney winning the popular vote by one percentage point. The NBC News/Wall Street Journal/Marist College poll incorrectly predicted that Obama would win North Carolina, and the CBS/Quinnipiac University poll incorrectly showed Obama losing Colorado.

In 2008, after accurately predicting the presidential result in 49 of the 50 states and in all 35 US Senate races, he was given "soothing credibility with Hollywood faithful", according to industry magazine the Hollywood Reporter.

"There's a blog that shows Obama winning?" Michelle Pfeiffer asked when told about the site.

"Then I'm reading it every day. I'm hanging on every word."

John Sides, a political scientist at George Washington University, said President Obama's victory "shows us that we can use systematic data - economic data, polling data - to separate momentum from no-mentum, to dispense with the gaseous emanations of pundits' 'guts', and ultimately to forecast the winner."

In the wake of this week's election result, the only non-polling comment made by Silver - who described himself on Twitter as a sports/politics/food geek - was a plug for one of his books, which experienced a spike in sales.

The post was re-tweeted more than 6500 times and "favourited" almost 2450 times.

Sales of the 544-page book, The Signal and the Noise: Why So Many Predictions Fail - But Some Don't rose by 800 on - ©The Daily Telegraph, additional reporting by Reuters


If you have an opinion you would like to share on this article, please send us an e-mail to the Times LIVE iLIVE team. In the mean time, click here to view the Times LIVE iLIVE section.