Throughout this election season, politicians and pundits alike have been focused on the subject of polling data. Whether at the national or state levels, the accuracy and reliability of polling data has itself become a subject of heated debate, with some viewing polling numbers as an accurate tool to gauge the state of a race, and others dismissing them entirely as false and irrelevant numerical distractions. Much of this distrust was sparked after the projections made in 2016 incorrectly predicted that Hillary Clinton would emerge as the certain winner. With the 2020 election on our doorstep, will the same mistakes be made?
What are the polls saying now?
As we approach the final days before the election, the Real Clear Politics polling average puts Joe Biden ahead of Donald Trump in national polls, with a 7.9% lead over the incumbent at 51.4% to 43.5%.
Despite complaints from several prominent Democrats, the president is not elected by popular vote, meaning that state polling data is far more consequential when it comes to predicting the actual election result. This has placed the polling data from 12 key battleground states under heavy scrutiny. Trump’s most likely path to victory requires him carrying 7 of those states: Florida, North Carolina, Arizona, Pennsylvania, Wisconsin, and Michigan.
Winning those 7 states would get Trump to 272 electoral votes, but there are plenty of other combinations that lead to victory.
Seen by many as an absolute must-win state for Trump, the polling data has been erratic since late August. After enjoying a strong 6.2% lead at the end of July, Biden’s advantage over Trump has shrunk to just 1.2%, with Trump even briefly overtaking Biden on October 26th by just 0.4%.
In 2016, Trump won Florida by 1.2%.
North Carolina’s polling data looks like a rollercoaster for both candidates. In the past few months, Trump and Biden have each temporarily taken the lead. Biden has outmatched Trump in this back-and-forth, but the current polls suggest this race will continue to remain incredibly tight, with Biden currently leading Trump by just 1.2%.
In 2016, Trump won North Carolina by 3.6%.
With a few exceptions, Biden has enjoyed a healthy lead over Trump in the average poll data since May. However, his sizable advantage recently vanished, placing each candidate at 47% on October 28th, and handing Trump a narrow lead of 0.6% on October 29th. This marks a surprising change in affairs for Biden, who held a 5.7% advantage over Trump as recently as September 9th.
In 2016, Trump won Arizona by 3.5%.
The polling data for another important state, Pennsylvania, has been less turbulent than others. Biden’s lead has been cut to just 3.6% in recent days, with Biden receiving 49.5% compared to Trump’s 45.9%.
In 2016, Trump won Pennsylvania by 0.7%.
Current polls are looking less promising for Trump in Wisconsin, with Biden extending his lead in recent days to 6.4%, with 50.3% in favor of Biden and just 43.9% for Trump. Days earlier on October 20th, Trump managed closed the gap to 4.6%, but the polling data soon moved in the wrong direction for him.
In 2016, Trump won Wisconsin by 0.7%.
Another common battleground state is Michigan. Since late September, it seemed like Biden and Trump’s poll results were diverging, with Biden holding a massive 9% lead on October 26th. Since then, Trump has been able to close the gap to 6.5%, which still leaves a large obstacle for the president to overcome.
In 2016, Trump won Michigan by 0.3%.
In July, Trump was enjoying a healthy 3% lead over Biden in the long-time Republican stronghold of Georgia. However, Biden rapidly closed the gap in late September, triggering a back-and-forth battle in the polls. Currently, Biden seems to be jumping ahead, establishing a narrow 0.8% lead over Trump.
In 2016, Trump won Georgia by an impressive 5.1%.
The state of Minnesota, with Minneapolis experiencing significant unrest after the death of George Floyd, hasn’t voted for a Republican president since 1972. In late July, Biden had a huge lead over Trump of 11.4%. In recent days, however, Trump has rapidly closed the gap, with 4.7% now separating the two candidates.
In 2016, Clinton won Minnesota by 1.5%.
As the Daily Wire reported in 2016, “a vast majority of the key polls were not just wrong, they were humiliatingly wrong.” As a result of these failures, we now see a widespread level of distrust toward the current poll numbers, particularly the ones which seem to mirror the same “clear path to victory” for Biden that was deemed certain in 2016.
On the morning of the election four years ago, Real Clear Politics’ polling average gave Hillary Clinton a 3.3% national lead over Donald Trump. Clinton was projected to win both the popular vote and the Electoral College. In reality, Clinton’s national lead was only 2.2%, and she lost the Electoral College 227 to 304.
The drastic inaccuracies in Electoral College predictions was a product of flawed state polling. Pollsters projected that Clinton would win Michigan, Pennsylvania, and Wisconsin by margins of 3.4%, 1.9% and a whopping 6.5% respectively. Trump went on to win each of these states by small margins (under 1%), with the pollsters missing the mark by 3.7% for Michigan, 2.6% for Pennsylvania, and an unbelievable 7.2% for Wisconsin.
Across the twelve battleground states of Nevada, Minnesota, Michigan, Pennsylvania, Wisconsin, Florida, Arizona, Georgia, North Carolina, Ohio, Iowa and Texas, the polls missed the actual result by an average of 3.2%.
Can we trust the polls this time?
In the aftermath of the 2016 election, there has been an understandable level of wariness when it comes to polling data for 2020. Many simply do not believe that Joe Biden’s lead — both nationally and within key battleground states — holds any material value, especially when combined with a clear gulf in enthusiasm for either candidate.
When analyzing “what happened in 2016,” The Washington Post reported, “many of the best-quality state polls still carry a margin of sampling error of three to five percentage points.” They went on to raise the issues pollsters typically face.
“Some types of people may be more likely to answer a survey than others, which pollsters work to correct for in the way they draw samples and by weighting samples to match population demographics or political characteristics.”
In addition to sampling errors, they also discussed the impact last-minute voting decisions can have on an election. They then moved on to describe “how pollsters are adjusting,” with two strategies being employed. The first is to conduct more polls in order to “contribute to greater precision overall.” The second is to use polls which “appear to be weighting samples by educational attainment,” apparently hoping to address the strong correlation demonstrated in 2016 “between education and support for Trump and Clinton in key states, with Trump winning by wide margins among White voters with some college or less.”
Of course, only time will tell whether this will make any difference with regards to the accuracy of the final projections. However, these adjustments fail to acknowledge the foundational cause of the polling inaccuracies in 2016, which was the failure to adequately consider Trump’s “silent voters.” The overt hatred expressed towards both Trump and his supporters has unleashed a predictable hesitance to express one’s political feelings, leading to “shy” Trump voters who are missed because they decide not to share their true opinions.
Neither the addition of more polls, nor the inclusion of supposed “education adjustments” will do little to encourage someone to admit they plan on voting for Trump if they are afraid to do so, thereby implying that the polls may be inaccurate once again. If in two elections in a row we see a clear inability to adequately measure and analyze the opinion of potential voters, then we will have further evidence that the old system is completely broken, and pollsters must go back to the drawing board. In the world of Trump and his “shy voters”, the old math doesn’t work.
Ian Haworth is host of The Ian Haworth Show and The Truth in 60 Seconds. Follow him on Twitter at @ighaworth.
The views expressed in this opinion piece are the author’s own and do not necessarily represent those of The Daily Wire.