November 6th, 2018

‘This Is Not A Normal Election’: Have Pollsters Failed To Account For ‘Unusual’ Factors?

Facebook

Twitter

Mail

According to pollsters, Democrats are poised to wrest control of the House from Republicans, who’ve managed to hold it for 8 years, while Republicans are about as likely to maintain their majority in the Senate, possibly even picking up a seat or two. But have the pollsters accounted for what analysts are calling a particularly “unusual” election, which features historic early voting turnout and what many believe will be unprecedented numbers of “non-usual” voters?

“This is not a normal election,” voting analyst and University of Florida Professor Michael McDonald told Politico in reference to voter turnout. “The best guess is that we’re looking at some sort of hybrid midterm/presidential election.”

So what’s so unusual? First, the early voting numbers, which have already reached some 36 million and are projected by pollster Tom Bonier to likely surpass 40 million when they’ve all been collected. In 2014, a total of 27.2 million people voted early. McDonald predicts that every state could end up surpassing its 2014 early voting numbers this year.

If pollsters failed to calibrate for that massive surge in early voting, Politico underscores, they could end up with results that diverge significantly from their predictions:

Early voters in three states — Texas, Nevada and Arizona — have already surpassed total turnout in the last midterm election, McDonald’s data shows, and more states will blow past their normal non-presidential turnout with just a handful more votes on Election Day. The high voting rates have transformed expectations about who will show up in the midterms — and they could inspire results that diverge from any pre-election polls that did not reckon with this year’s unusually high enthusiasm.

McDonald predicts that a total of about 105.5 million people will vote this year, which would be over 20 million more people than in 2014 and about 30 million fewer than voted in 2016.

Another unusual aspect of this election, which Bonier’s group TargetSmart has tracked, is the higher than normal participation of “non-usual” voters, including first-time voters and young voters, the impact of which is yet unclear.

Despite the unprecedented factors, Bonier told Politico that pollsters have “all made a very safe assumption that 2018 will look nothing like 2014,” and have attempted to calibrate accordingly. If they haven’t, we might end up with another November 2016.

As for what the pollsters project will happen today, analysis of all the most recent polling by Nate Silver’s FiveThirtyEight gives Republicans a 4 in 5 chance (81%) of maintaining control of the Senate and Democrats a 7 in 8 chance (88.1%) of winning control of the House. Below is The Daily Wire’s summary of Real Clear Politics’ averages for both the Senate and House as of the morning of the election:

For the last few weeks, Real Clear Politics‘ average of the key polls has given Republicans 50 “safe” Senate seats and narrow leads in two of the seven “tossup” races (Missouri and Nevada). On election morning, however, RCP only gives Republicans 49 “safe” seats, having shifted Tennessee into the “toss-ups” column (the Republican leads by 5 points). Last week, RCP gave Democrats 44 “safe” seats; that number has dwindled by one (West Virginia) to just 43. […]
Real Clear Politics‘ poll averages on the morning of the election show Democrats with 203 seats that are “safely” blue, including 15 that are “likely” to go their way and 15 that “lean” Democrat. Republicans have 194 “safe” seats (20 likely and 25 leaning). Last week, RCP gave Republicans 196 “safe” seats and Democrats 203. If RCP’s “safe” predictions hold up, Democrats only need 15 of the 38 tossup seats to attain the 218-majority in the House. Republicans, meanwhile, need to win 24. Thirty-three of the 38 tossups are currently held by Republicans.