In light of the GOP nominee’s claims that the polls are intentionally skewed towards democrats and thus, corrupt, we decided to see how accurate the polls have been throughout the years.
Most major polls have a 3-5% point margin of error per candidate, which means that any given poll could potentially be off by a maximum of 10%. While this is unlikely, it is possible and it is clearly significant. But there is an easy way to account for the margins of error in individual polls; just look at the average. Real Clear Politics (RCP) features a running average of all major polls. They are currently showing Clinton at +7.
Let’s look at or the RCP average has faired over the last three elections:
In 2012 the RCP Avg. = Obama +0.7
2012 election results = Obama +3.9
Obama outperformed the polls: +3.2.
In 2008 the RCP Avg. = Obama +7.6
2008 election results = Obama +7.3
Obama slightly underperformed: -.3.
In 2004 the RCP Avg. = Bush +1.5
2004 election results = Bush +2.4
Bush slightly outperformed the polls: +.9.
So if we take the RCP average going into the prior three presidential races, the numbers were only off by 3.2, .3 and .9. And in no instance did this margin of error account for a significant swing for any candidate.
No evidence of corruption so far.
Let’s look into one of the nation’s oldest, and most well-known polling companies: Gallup.
If we go back to the Kennedy v. Nixon election of 1960 and work forward, we see two times¹ when Gallup picked the wrong president. The first time was in 1976, in a very close race between Jimmy Carter and Gerald Ford. Final Gallup polls had Ford at 49% and Carter at 48%. But Carter ended up beating Ford, 50.1% to 48.1%. The difference is well within the margin of error, so it hardly points to Gallup’s numbers being untrustworthy. The second time Gallup whiffed was 2012. Final numbers had Romney up on Obama 50% to 49%. Obama ended up winning 51% to 47%. Even though they were off in the last election, the difference was within the margin or error.
1. A third misfire worth noting is the 2000 election. Gallup had Bush winning, but he ended up losing the popular vote. Again, their numbers were very close to the final outcome and well within the margin or error.
So over the last 14 presidential elections, Gallup has missed two, and one of them was too close to call. What we can take from this is that, if the race is really tight, it can swing either way on election day. But when the race isn’t close, when a candidate is up by more than 5 points, the other candidate has never been able to make up enough ground on election day.
In 1980, Ronald Reagan was down 8 pts to Jimmy Carter in a late October Gallup poll. Following their only debate on October 28th, the tide began to swing in The Gipper’s direction. By election day, Reagan had turned an 8 pt deficit into a 3pt lead. He went on to win the election by nearly 9 pts. Again the margin of victory was off, but this only serves to prove the malleability of the poll itself. When a shift occurs, it’s directly represented in the numbers.
So, by in large, the polls aren’t skewed. A few certainly lean a bit, but if you work off an average, the outliers negate each other. Polls are not perfect; but they are reliable, representative samples of how our nation is feeling leading up to an election.
Trump’s assertions are desperate, and they fall in line with the, I don’t lose, other people cheat narrative he likes to unfurl whenever he thinks he’s about to lose. It’s never him, it’s always someone else. But any way you spin these numbers, Trump is in big trouble. Even though this very well may be the most volatile and unpredictable race we’ve ever seen, Trump has a ton of ground to make up to shift this thing. And a Reagan-like comeback seems like a pipe dream at this point. For one thing, Reagan was able to swing popular opinion after a late stage debate with Carter, while Trump and Clinton have already had two debates and neither resulted in a Trump bump—quite the opposite actually. And in recent weeks Trump’s numbers have been plunging not ascending—he’s trending in the wrong direction.
Of course, even if many of these polls end up being dead-on, Trump won’t accept it. Because it’s not only the polling that is rigged, it’s the voting itself, it’s the entire country, it’s nation-wide collusion!
It’s SNL! It’s the New York Times! It’s the Washington Post! It’s Fox News! It’s People Magazine! It’s CNN! It’s PBS! It’s the debate moderators!
Only, it’s not.
It’s the American people. They just don’t like you very much, Donald.
They don’t love your opponent either. But they like her more than you.
by Jesse Mechanic
Jesse Mechanic is the editor in chief of The Overgrown.