This is an interesting review of the election dynamics. The point made is that there are too many unknowns to make a prediction. This is from The Bulwark
We are now past another of the big campaign milestones: Labor Day.
How does the race look?
Pretty much like it's looked the entire way:
![[IMG]](https://mcusercontent.com/41df14e6667a85d0f6e4a4f5e/images/ce63f85f-8fd0-47f5-a0a3-57a3a1753ba2.png)
Let's see how this compares to the polling in 2016:
![[IMG]](https://mcusercontent.com/41df14e6667a85d0f6e4a4f5e/images/f0871b55-1bac-4e6b-bbab-9697e2b23cd3.jpg)
A few points to note:
- On September 8, 2016, Hillary Clinton was +3.3
- Today, Biden is +7.5, so his lead is more than double what Clinton's was at this point.
- Relative to 2020, the 2016 race was more volatile. There were moments when Clinton and Trump were tied; moments when Trump was in the lead; moments when Clinton's lead was large (+8), and moments when Clinton's lead was within the margin of error.
- To date, Trump has never been closer than -3.4 against Biden.
- On September 10, Biden will have been over the 50 percent mark for a full month. Clinton was never over 50 percent, or even very close to it.
Now, what does that all mean?
It means this: If the election were held under normal circumstances then the most likely outcome would be a Biden victory of somewhere between +5 and +7 in the popular vote.
What's more, the confidence level on that likelihood would be as high (or higher) than it's been at this point in any recent election.
But there are two problems:
(1) The election will not be held under normal circumstances. I cannot emphasize this enough: It is almost impossible to make useful assumptions about what the pandemic will do to voting. Here is a chart showing the raw number of total votes in presidential elections dating back to 1900.
![[IMG]](https://mcusercontent.com/41df14e6667a85d0f6e4a4f5e/images/166c29f6-f960-4b46-b494-5158264b65b7.jpg)
Any model of the eventual outcome relies on having an educated guess on the range of total votes to be cast. Post
9/11, our election turnout have ranged from 121 million to 130 million votes.
In the pandemic environment do you think the total turnout is likely to be closer to 100 million? Or 140 million?
Nobody knows.
And if you can't even get a baseline on raw vote totals, then everything else in a model is going to be a crapshoot.
(2) The Electoral College advantage for Republicans has grown to such a degree that even a +5 popular vote margin for John Biden leaves Trump with a
10 percent chance of victory in the EC.
All of which is to say that we are in a paradox. We know what the mood of the country is. We know what people think.
But we have no idea who will be elected president 55 days from now.
Click to expand...