On November 9th 2016 the political establishment, media elites and Washington insiders were in a state of shock. What had just happened was the greatest Political revolution in American history. An outsider businessman and reality TV star, ousting the candidate once thought to be undefeatable- Secretary of State Hillary Clinton.
Donald Trump broke all known wisdom, conventions, and rules of American politics. He insulted Mexicans, Muslims, John McCain, the Khan family- the list goes on.
However, if you really look at the choice, it was always entirely conceivable that Trump could win. The crowds Donald Trump attracted to his campaign rallies were huge, often around 25,000 to 30,000 people attending, with thousands waiting outside the venues. Hillary struggled to enthuse anybody (even her own supporters were pretty low energy and lacklustre). It was well reported during the final days of the campaign that Hillary’s running mate, Senator Tim Kaine, drew a crowd of only 30 people in the crucial battleground state of Florida (while Trump drew crowds of around 20,000 in Naples, Florida).
If the media went out on the campaign trail to sense the mood on the ground, they would have been more prepared for the election result. Unfortunately for them, they decided to isolate themselves in a liberal New York City bubble. The news networks who were in bed with the Clinton campaign (MSNBC, CNN, ABC, NBC, NY Times, Washington Post, Huffington Post) produced phoney polls designed to suppress voter turnout in order to favour Hillary Clinton.
The perfect example of a fixed poll is this one produced by YouGov a few days before the election. According to this poll, Hillary Clinton had a comfortable lead in Michigan and it was likely to vote Democratic. However the actual results from Michigan show Trump pulled ahead in Michigan by over 20,000 votes. How could the polls be so wrong, so close to election day?
The answer is that, the pollsters used various tactics to skew the outcome of their polls. For example, many polls were conducted in Spanish to ensure that more Latinos were sampled (favouring Hillary Clinton). Other polls sampled more women than men, when it is well known women tended towards Clinton.
Perhaps the biggest miscalculation of the Pollsters however, is that they fundamentally misread the American electorate. America is still a centre-right country. Polls though, sampled more democrats than republicans. In an ABC Poll for example, the sample was comprised of 27% Republicans, 36% Democrats and 37% independents. After Hillary obviously won in this poll, ABC then went on to report to the world that Hillary had a 12 point lead over Donald Trump. Crazy!
The media found the outcome they wanted and fixed the facts to meet their outcome. They wanted Hillary to win, so they convinced themselves that she was winning. The pollsters should not have assumed that there are more Democrats in America than Republicans, after all in small town America, Donald Trump inspired unprecedented turnout. In fact, Trump won the white working class vote by over 30 points- a higher margin of victory than Ronald Reagan had.
Here’s the bottom line, you cant shut yourself off from reality and hope and pray that its not happening. Trump was happening. The establishment protected itself, but in the process had destroyed its own reputation. Some of the most accurate polls were from independent organisations. A Trafalgar Group Poll for Florida showed Trump leading in Florida by 4 points. While this is not absolutely accurate, it was one of the only polling organisations that projected a Trump win in Florida- and it was accurate because it sampled an equal number of Republicans and Democrats. If ever you need proof of what I have just alleged, look at the pollsters methodology, if they don’t include methodology, take it with a pinch of salt. If they do include a methodology but have made great assumptions, then again, take it with a pinch of salt.
The state of Pennsylvania highlighted the national trends in the US election. The polls assumed a Clinton victory here, believing that the margin of victory Barrack Obama got out of Philadelphia county would be matched by the Clinton campaign. This was a great miscalculation. The idea that Hillary would be able to bring out the coalition of minorities to get as many raw votes as Obama did, is crazy. Obama won Philadelphia by 490,000 votes, where as Hillary won it by 450,000 votes. She was down by 40,000 votes in Philadelphia, allowing Donald Trump to run up the vote in rural areas and working class strongholds. Look at Lackawanna county, where Scranton Pennsylvania is. This is a county where Barrack Obama won by 30 points. In this election, Hillary won it by only 4 points. White, Blue collar workers came out in droves for Trump, while Hillary didn’t get the turnout the polls the assumed she would get.
The reality is that polls are rarely 100% accurate. They are after all just a snapshot of societies opinons at one point in time. Opinions change, and often times some voters are shy and don’t want to speak to pollsters, or they are embarresed of their voting choices so they lie to appear more socially acceptable. If the polling industry is to be trusted again, they have to start opening their eyes to the reality of the electorate—not just isolating themselves in New York City.