A few days ago a YouGov poll came out showing Trump Trailing Biden by 10%. It had Biden at 52% and Trump at 42%. At first I was surprised because YouGov was actually one of the few pollsters that was pretty close to predicting the popular vote in 2016. Their poll was only 2% off from the actual popular vote, and they were fairly accurate with polling independent voters. Just look at the break down of their 2016 poll.
This was the actual 2016 exit polling by party. The image below shows how people really voted in the 2016 election.
As you can tell, they were pretty accurate across the board. They almost had a perfect prediction with republican and democrat voters. They also did a pretty good job of predicting independent voters. Overall, YouGov was one of the best pollsters in 2016. They also did a pretty good job of polling in the 2018 midterms.
However, the same cannot be said about the 2020 election cycle. YouGov, whether they’re doing it on purpose or not, has been throwing their polls off. YouGov and many other pollsters have been oversampling democrat voters and under sampling republican voters in their polls. Take a look at some data from the 2016 and 2018 elections. In 2016, 36% of all the voters that voted nationwide were registered democrats, 33% were registered republicans, and 30% were unaffiliated with any party (AKA independent). In 2018, 37% of voters were democrats, 33% were republican, and 30% were independent.
As you can see, voter turnout didn’t change much between these two elections, so you could probably bet that it wouldn’t change much for the 2020 elections.
This brings us back to the YouGov poll from yesterday. They had Trump at 42% and Biden at 52%. When I saw this I was shocked because typically they do good polling. However, when I looked into the poll, I saw why it had Biden so far ahead. They over sampled democrats and under sampled republicans by a lot. Here’s their sampling from their poll.
|Registered Democrats||Registered Republicans||Independents|
So let’s get this straight. They over sampled democrat voters by about 5%. They under sampled republican voters by 3%, and they under sampled independent voters by 3%.
It’s no wonder the poll shows Biden so far ahead. What’s even better is that Trump is leading with independent voters just like he did in 2016. In this poll, 47% of independent voter said they would vote for Trump with only 37% saying they would vote for Biden. Here’s the break down by party of the poll.
When you take these numbers and set them to the correct levels of republicans, democrats, and independents, Trump isn’t that far behind. Using the 2018 midterm model and the crosstabs from the poll above, Trump would actually be at 46% and Biden at 48%. These numbers should look familiar. In 2016, Trump lost the popular vote with Hillary Clinton at 48% and Trump at 46%.
Trump has the exact same level of support that he had in 2016. However, almost every single pollster out there shows him behind by 5%, 10% and even 15%. Don’t be fooled by this. The crony media and pundits could be in for a rude awakening on election night.
At this point, I think that Biden and Trump both have a shot at winning this election. However, I wouldn’t rule out the possibility of a landslide victory for Trump. We’ve still got a long way to go before election night, and a lot could happen. These polls are meant to demoralize republicans. These pollsters want Trump voters to think that Biden has the election in the bag. They want Trump voters to think that it’s over for the President already. Don’t believe them. The polls are even worse than they were in 2016, so don’t get demoralized. Get out and vote for Trump this November, and prove them wrong.