Not One Major Poll or Analyst Predicted Trump Defeating Clinton

From Reuters to Nate Silver, from the New York Times to the Washington Post, a great majority of polls had indicated a Clinton victory.

US President-elect Donald Trump greets supporters during his election night rally in Manhattan, New York, US, November 9, 2016. Credit: Reuters/Mike Segar

US President-elect Donald Trump greets supporters during his election night rally in Manhattan, New York, US, November 9, 2016. Credit: Reuters/Mike Segar

“Polling was wrong, everyone was wrong, we were wrong,” said a commentator on CNN on Wednesday afternoon (IST). And that pretty much sums it up.

Whatever else you may say for it, 2016 has been a year of unexpected election results. Analysts, polls and political commentators got it wrong this June with their projections of how the Brexit vote would turn out – the outcome even surprised some of the staunchest supporters of the UK leaving the EU.

And today we see it again – after a nail-biting counting period, Donald Trump is the new president of the US. And once again, large sections of the world who’ve been following political analysis are left surprised (or shocked) and confused.

The Wire looks at all those who confidently predicted a win for Hillary Clinton, many even less than 24 hours ago.

Nate Silver, editor-in-chief of FiveThirtyEight and a widely respected election forecaster, was reserved even as he said a Clinton victory was the most plausible outcome. His website was slightly less precocious, saying that there was more than a 71% chance of a Clinton win.

Other analysts, polls and news agencies were far less reserved in their predictions. Just a few hours before polling, a Reuters/Ipsos poll said there was a 90% chance of a Clinton victory. An Upshot poll published in the New York Times said Clinton had an 85% chance of winning. The Washington Post on November 7 said “the overall map still clearly favors Hillary Clinton”. Emerson College Polling predicted a victory for Clinton, giving her 323 electoral votes and saying she would win in all the swing states except Ohio and Iowa – and while they were right about Ohio and Iowa, they were wrong about pretty much everything else. The Huffington Post presidential forecast model predicted the same number of electoral wins for Clinton, adding that she had a whopping 98.2% chance of winning. The Los Angeles Times gave her 352 electoral votes on their final electoral map, though a poll they conducted four months ago showed Trump leading.

Not one major poll said a Trump victory was in the offing. Wall Street analysts also predicted a Clinton victory.

Data analyst and founder of PredictWise David Rothschild was also pretty confident less than 24 hours ago, though his number changed continuously after counting started. He had said Clinton had an 89% chance of winning.

Commentators affiliated with both major parties were also pretty off, on the whole. GOP pollster Frank Luntz tweeted before counting began:

GOP strategist John Weaver was also pretty confident of Clinton being elected. “I believe she’s going to win in an electoral landslide and be the most unpopular president in electoral history, which is quite the paradox,” Weaver told David Axelrod on ‘The Axe Files’ podcast on election day.

Buffy Wicks, a Democratic strategist, wrote a confident article in Time Magazine on November 8, announcing the inevitability of Clinton’s victory. “I hope the analysts and pundits recognize this campaign for what it truly is—a historic achievement of grassroots mobilization in a time of unparalleled political polarization,” she wrote.

And it wasn’t just those in the US who called it for Clinton. A few days ago, the Times of India published an article by Bryan Cranston titled ‘A Trump victory is (nearly) mathematically impossible’. “Despite what the public polls suggest, and even with her latest email scandal, the election is Clinton’s to lose, and it appears mathematically unlikely that she will,” Cranston wrote.

Economist Surjit Bhalla also announced a Clinton victory on Tuesday. “The long wait will soon be over. As everyone has noted, this is not an ordinary election. The future of Western civilisation (literally) depends on the outcome. We are confident of our forecast; if we are wrong as in Trump winning, or Clinton winning by a narrow 2-3 per cent margin, we will be very unhappy, and not just for making a wrong forecast,” Bhalla wrote in Indian Express. He had predicted 52% of the vote going to Clinton.

A newspaper in Kenya, among others, also confidently predicted a Clinton win.

A group of scientists at the University of Utah analysed tweet “sentiment analysis” to guage the sentiments expressed in 1.6 million geo-tagged tweets, DNA reported. This method has proved to be no more useful. “Based on the number of positive tweets posted since June towards each party, the computer model predicts that Democratic Party nominee Clinton will win the US presidential election,” a DNA report on the research said on November 5.

On a lighter note, it wasn’t just those going by models and numbers who were off in their calculations.

Boots the “psychic” Scottish goat had no luck either. But Geda the Chinese monkey was on point.

Maybe 2017 will be different – not only a year for less “shocking” elections results and less confidently erroneous predictions, but also a year for getting closer to understanding a polarised global society.