Be skeptical of the polls: Wait for the results to celebrate


Dante Harrold, Staff Writer

Everyday I go to polling aggregate sites RealClearPolitics and FiveThirtyEight to see how the polls are progressing. It’s become the first thing I do in the morning before brushing my teeth, taking a shower or eating, I turn on my phone and see if there’s been any change in polling numbers for Presidential candidate Joseph Biden or for President Donald Trump.

I do this despite knowing for a fact that doing this will never soothe my nerves even a little bit. I cannot confidently say President Trump will be re-elected nor can I confidently say that Biden will win and become America’s 46th President. I can only confidently say that the polls can be wrong.

Pollsters acknowledge the fact that polls will be wrong to a degree. They take this into account when looking at their margin of error because they cannot reasonably expect their polling to be 100% accurate when compared to the final numbers on Election Day.

It may sound odd to many, but the predictions of the 2016 election were just as wrong as the 2012 election. The RealClearPolitics polling average for former President Barack Obama had him ahead nationally by just 0.7 percentage points, but he went on to win the popular vote by 3.9 percentage points.

At the state level, the error in certain states had Obama underperforming by a wider margin 

RCP had Obama winning Michigan by 4 points and in the end he won the battle-ground state by 9.5 points.

And though people were shocked that Hillary Clinton lost the election in 2016, FiveThirtyEight had numerous articles showing that the race was much closer than pundits would have you believe at the time. In the article “Trump Is Just A Normal Polling Error Behind Clinton” by Harry Enten, he explains that polls can be extremely accurate, but can also be way off the mark of the actual results. Many pollsters had Clinton winning by a wide margin, which obviously wasn’t the case come Election Day. 

”Polls of the November 2016 presidential election were about as accurate as polls of presidential elections have been on average since 1972.” said Nate Silver, the founder of FiveThirtyEight. 

He went on to explain that is the nature of polling, that they should be taken with a grain of salt. 

“While some will be spot-on, many others will be off by 5 or 10 points or even more — and this will be perfectly normal”

As of Nov. 2,  FiveThirtyEight estimates through simulation that Biden will win 89/100 times. The polls will be wrong and whether or not they will be wrong in Biden’s favor or President Trump’s will be revealed on Election Day.