Toro
Diamond Member
- Sep 29, 2005
- 109,797
- 50,574
First, take the 10 swing states. They are
AZ FL GA IA MI NV NC OH PA TX WI
These 10 states account for 185 electoral college votes.
Trump won all except NV, or 179 EC votes.
Trump won 306 EC votes, Hillary won 232.
Assuming all the non-swing state votes do not flip, Trump starts with 127 (306-179=127) EC votes and Biden starts with 226.
Using the final polling numbers from 538.com, we can see Clinton's final polling numbers and compare it to her actual vote. The yellow column is the difference.
You can see that in most states, Clinton underperformed her polls by 1.2% on average. (You can do the math.)
Remember that "Difference" yellow column.
Doing the same thing for Trump, we can see that he outperformed his polls in almost all states, and on average by 2.0%.
Again, remember that yellow column.
Let's look at the polls for Biden and Trump now.
Now let's apply the difference between the polls and the actual vote from 2016, i.e. the yellow column, and apply that to the current polls for each candidate.
Biden
Trump
Now let's compare Biden and Trump adjusted for the 2016 polling error
Biden would thus win 334 EC votes compared to Trump winning 204.
This means that for Trump to win, the polling error would need to be bigger than 2016.
Let's say you adjust Trump's vote up 1%
It would be a tie. Biden would have 269 (226+43) and so would Trump (127+142).
This is where Trump totally blowing it in the suburbs would cost him.
Our model assumes everything else stays the same. But in NE-02, which is essentially Omaha, Trump is getting trounced.
The Republicans won that by 9.4% in 2016, but are trailing by 4.5% today, or a 13.9% swing. That's high but not uncommon in the suburbs.
If that one EC vote were to swing to Biden, then he'd win the election.
AZ FL GA IA MI NV NC OH PA TX WI
These 10 states account for 185 electoral college votes.
Trump won all except NV, or 179 EC votes.
Trump won 306 EC votes, Hillary won 232.
Assuming all the non-swing state votes do not flip, Trump starts with 127 (306-179=127) EC votes and Biden starts with 226.
Using the final polling numbers from 538.com, we can see Clinton's final polling numbers and compare it to her actual vote. The yellow column is the difference.
You can see that in most states, Clinton underperformed her polls by 1.2% on average. (You can do the math.)
Remember that "Difference" yellow column.
Doing the same thing for Trump, we can see that he outperformed his polls in almost all states, and on average by 2.0%.
Again, remember that yellow column.
Let's look at the polls for Biden and Trump now.
Now let's apply the difference between the polls and the actual vote from 2016, i.e. the yellow column, and apply that to the current polls for each candidate.
Biden
Trump
Now let's compare Biden and Trump adjusted for the 2016 polling error
Biden would thus win 334 EC votes compared to Trump winning 204.
This means that for Trump to win, the polling error would need to be bigger than 2016.
Let's say you adjust Trump's vote up 1%
It would be a tie. Biden would have 269 (226+43) and so would Trump (127+142).
This is where Trump totally blowing it in the suburbs would cost him.
Our model assumes everything else stays the same. But in NE-02, which is essentially Omaha, Trump is getting trounced.
The Republicans won that by 9.4% in 2016, but are trailing by 4.5% today, or a 13.9% swing. That's high but not uncommon in the suburbs.
If that one EC vote were to swing to Biden, then he'd win the election.