It still isn't clear who will be declared the winner of the US presidential election, but even Wednesday night we could say that once again, the polls had misled the Americans. On the eve of the election, it looked like Democratic candidate Joe Biden had a big, solid lead, and was on his way to a large victory. But the moment it turned out that Florida was going for Trump, it was clear that the polls had been far from reality and, like in 2016, had not correctly assessed the Republican president's strength.
If we look at all the votes counted thus far, it seems as if the polls from four years ago were more accurate for today.
Follow Israel Hayom on Facebook and Twitter
On the eve of the 2016 election, most pollsters were pointing to a small lead for Democratic candidate Hillary Clinton, who was ahead by an average of 3.2%. In the end, Clinton lost in electoral votes, but won the popular vote by a margin similar to what the polls showed of 2.1%. This time, Biden's projected lead in the polls stood at 7-8%. Last night, there were millions of votes that were still uncounted, but Biden's lead nationwide was much smaller, at some 2%.
The apparently lack of accuracy applies not only the national average but also in the very close races in a number of key states. In Florida, for example, most averages gave Biden a lead of 1-2%. With 96% of the votes counted on Wednesday, Trump won the state by 3.5%, 5% off the average in the polls. In Ohio, another state where the president won, the poll averages gave him a lead of 1%. With 96% of votes counted, he won with more than an 8% margin.
The errors of the polls were just as big for the Rust Belt. In Wisconsin, the former vice president was polling ahead by some 7%. As of Wednesday night, it looked like the Democratic candidate has won, but by a tiny 0.6% only, far from the lead the polls had promised. In Michigan, Biden's lead in the polls averaged 4.2%, but on Wednesday night, with 96% of the votes counter, he barely had a 0.9% lead. The situation in Pennsylvania, the third Rust Belt state, was still unclear on Wednesday, but polls gave Biden a lead of only 1.2% there.
Despite the problems, pollster Nate Silver, editor of the well-regarded site 538, rejected claims made against him and his colleagues, tweeting: "If a forecast says that Biden is favored because he could survive a 2016-style (~3 point) polling error when Clinton couldn't, and you get that polling error and he indeed (probably) survives, it was fairly informative?"
Subscribe to Israel Hayom's daily newsletter and never miss our top stories!
