trump biden poll election enten nr vpx_00000000
Enten: This is the latest sign that Trump is in trouble
02:38 - Source: CNN
CNN  — 

Poll of the week: A new Franklin & Marshall College poll from Pennsylvania finds former Vice President Joe Biden at 50% to President Donald Trump’s 41%.

This is just the latest Pennsylvania poll to show Biden with an advantage. A mid-July Fox News poll put Biden’s margin at 11 points.

What’s the point: Trump won the 2016 election in large part because he was able to break through the big “blue” wall in Michigan, Pennsylvania and Wisconsin. He did so on the backs of White voters, and in particular White voters without a college degree. This led to those infamous memes of reporters seeking out voters in diners across America.

But today, Biden leads Trump in these Great Lake (or Rust Belt) battlegrounds because he’s eating into Trump’s margins among those very groups. These gains have big implications for the electoral college because they suggest Biden’s easiest path to victory may be through these three states.

Biden has clear leads in an average of the last three CNN approved polls in the states of Michigan, Pennsylvania and Wisconsin. On average, Biden is up by 9 points in Michigan, 11 points in Pennsylvania and 10 points in Wisconsin.

Biden’s upward trajectory is because of vast improvements among White voters in a comparison of Biden’s standing in each poll to how Clinton did with them in an average of post-election estimates from the Cooperative Congressional Election Study and Center for American Progress.

Post-election estimates aren’t the most ideal comparison to pre-election polling, which would be accurate pre-election polling (something missing from these states in 2016). Like any poll (or poll average), these post-election estimates are subject to error.

Still, the size of Biden’s rise over Clinton in these states is clear enough that it’s well outside the range of any statistical anomaly.

  • Michigan: Trump leads by an average of 3 points among White voters. Four years ago, Trump won amongst these voters by 15 points.
  • Pennsylvania: Biden is ahead by 3 points with White voters. In 2016, Trump won them by 15 points.
  • Wisconsin: Biden’s up by 6 points with White voters. Last election, Trump took them by 7 points.

The key takeaway here isn’t the exact shift in each state (which is subject to a margin of error), but that what we’re seeing nationally in regards to White voters is occurring in the key Rust Belt states as well. Using the method nationally, Biden’s improved from Clinton’s 15-point loss to a mere 3-point deficit against Trump in the latest live interview polls. (This is quite similar to the movement seen using pre-election polls as our 2016 comparison point, which also shows a double-digit improvement for Biden.)

Obviously, the fact that Clinton underperformed her pre-election polls in these states should give us pause. Amazingly, however, these Rust Belt numbers among White voters are so good for Biden that he’s in a considerably better position than even the final high quality pre-election polling that looked rosy for Clinton in each state.

View 2020 presidential election polling

We can drill down further to examine non-college educated White voters as well. Not every poll provides that crosstab, so I combined the different states to get a large enough sample to compare. Across the six polls that did include a crosstab for Whites without a college degree, the trend is very clear.

Biden’s doing on average 12 points better among Whites without a college degree than Clinton did in the combined sample of Michigan, Pennsylvania and Wisconsin polls. He’s obviously still trailing among White voters without a college degree, but the margin on average is closer to 10 points in favor of Trump than 20 points.

Again, this is quite similar to what we’re witnessing nationally. Biden’s doing about 10 points better among Whites without a college degree than Clinton did in a comparison of current polling to pre-election polls from 2016. That’s why a model by The New York Times’ Nate Cohn projected similar overall results in these three states based off of the national data.

The fact that White voters and specifically White voters without a college degree in the Rust Belt are trending in the same direction as those groups nationally is a big deal because they make up a much larger chunk of the vote in the Rust Belt. White voters will likely make up north of 80% of voters in Michigan and Pennsylvania and closer to 90% in Wisconsin in 2020. Nationally, it’ll be closer to 70%.

White voters without a college degree will likely make up a majority of voters in Michigan, Pennsylvania and Wisconsin. It’ll be near 60% in Wisconsin. Nationally, Whites without a college degree will only be about 40% of the electorate.

In other words, the shift among Whites and White voters without a college degree we’re currently witnessing in the polls will have a greater impact on the outcome in these key Rust Belt battleground states.

Not surprisingly, these three states have swung harder in Biden’s direction than the average of national polls. At this point, there isn’t a large difference between the national and state polls in Michigan, Pennsylvania or Wisconsin. If a difference does exist, as it does depending on how you average the polls, it’s smaller than in 2016. That year Trump outran his national margin by 2 to 3 points in these Rust Belt battlegrounds.

That’s a big deal because Michigan, Pennsylvania and Wisconsin combined have 46 electoral votes. If you add those to the 232 in states Clinton won in 2016, you get to 278 electoral votes.

Biden wouldn’t need to win over any other Trump won states in order to win the election.

Were this trend to hold until the election, Biden’s most reliable path to get to 270 electoral votes is in the Rust Belt. Biden’s doing better in all three of these Great Lake states than he is doing in any of the other states Trump won in 2016.