Monday, October 26, 2020


I keep thinking about this chart, from the Upshot at The New York Times:

I keep staring at the column labeled "If Polls Are as Wrong as They Were in ... 2016." I don't think they will be as wrong -- most pollsters now know that they need to weight their polls for education, so I don't think we'll have as many uncounted Trump voters. But who knows?

If we do, notice how the election turns out. Joe Biden wins the popular vote by 6 points. Let's eliminate third-party candidates and say that's a 53%-47% victory. It would be the second-largest popular-vote win this century, behind only Barack Obama's 7-point win in 2008.

But Obama won the Electoral College by more than two to one. He got 365 electoral votes. John McCain got 173.

According to the chart above, in this scenario, Joe Biden won't win Wisconsin, Florida, North Carolina, Georgia, Texas (where a new Times poll out today says Biden is trailing), or Maine's second district. He'll win Minnesota and Pennsylvania, but barely.

The electoral vote under these circumstances? Biden 280, Trump 258. A squeaker.

This may happen despite a popular-vote margin of something like 9 million. That's what 53%-47% looks like if -- as some observers predict -- we have turnout of 150 million voters.

And it seems to me that if the polls are off the same way they were in 2016 but slightly more so, that could give Biden a 5-point popular-vote win -- a win by 7,500,000 votes -- while Trump wins Pennsylvania and Minnesota, and is thus reelected with 288 electoral votes. (Yes, Trump could win more electoral votes with a 5-point popular-vote loss than Biden would with a 6-point popular vote win.)

Even if this doesn't happen, it's outrageous that it could happen. This is not democracy. It's insanity.

No comments: