2020 Election “Prediction”

First, I have about as much confidence in this as I do in my March Madness bracket (which is not much)…This is submitted for fun and it is submitted with humility. Here goes!

Assumptions guiding my decisions:

  1. My make or break assumption is that the polls are not accurate…again. I doubt the pollsters learned nothing from 2016, but I do believe “social acceptability bias” is influencing the polls and that Trump support is undercounted. Therefore, I am going to put more weight on the Trafalgar Group polls than the RealClear or 538 averages. Trafalgar was one of the few firms to get it right in 2016. That said, one of the greatest mistakes strategists can make in war is that they fight the last war…they can only see the current conflict through that lens. I fully admit I may be making the same mistake here.
  2. Charisma matters. This research has me convinced that Trump voters are more enthusiastic and will therefore actually show up at the polls in greater numbers. I will take this into account when polling data is split or extremely close.
  3. Primary elections are a better predictor of turnout than polling, according to the Primary Model. This model has predicted 25/27 of past presidential elections correctly and a Trump win is forecasted due to Biden’s poor performances in the first few primaries. I will use this assumption to swing very close states to Trump.
  4. The big unknown is how the pandemic will impact the election. My assumption is that a majority of people (deep down) are actually more concerned about the economy than they are about COVID-19. This favors Trump, if true. 

I looked at the data for each state and applied these assumptions when it was close. I did this without looking at the electoral count until I was done. The results are depicted above.

RCP List of “Battleground” States

Florida = Trump

Trafalgar has Trump up by .04%. RCP has Biden up by 1 point. Trump won in 2016 and I think support from Cuban Americans and the information reflected in the above assumptions will ensure he holds.

Pennsylvania = Biden

Even Trafalgar has Biden up by 2.3 points. 538 has Biden up by 6 points. I thought Biden’s statements on oil would impact the polls more, but they don’t seem to have budged. 

Wisconsin = Biden

RCP average has Biden up 6 points. Trafalgar has Biden up 1.3%. 

North Carolina = Trump

Trafalgar has Trump up by 2.8. RCP has him up as well.

Michigan = Trump

Trafalgar has Trump up by .6 points. RCP average has Biden up by 5.1%. Michigan went red in 2016. Here is where the candidate charisma and voter enthusiasm come in to play. Trump wins both of those according to the linked sources, so I am throwing MI to Trump. This decision basically determines the outcome of my prediction.

Ohio = Trump 

RCP has Trump up by .2% and Trafalgar has him up by 5 points. 

Minnesota = Biden

Almost every poll has Biden way up there.

Iowa = Trump

RCP has Trump up with a 1.2% advantage. Well within the margin of error, but I’m betting on this assertion that enthusiasm for Biden has collapsed. 

Arizona = Trump

RCP has Biden up 1%. Trafalgar has Trump up by 3%. My instinct in AZ (my home state) is that Republicans will win close races.

Nevada = Biden

Even Trafalgar has Biden up by 2 points.

Texas = Trump

RCP has Trump up by 1.2%. If Beto couldn’t turn Texas, I don’t see Biden doing it. 

Georgia = Trump

RCP has Biden up by .4%. I will throw GA to Trump due to the above assumptions.

Virginia = Biden

It’s not even close.

New Hampshire = Biden

It’s not even close.

Maine = Biden

See above

Colorado = Biden

See above

New Mexico = Biden

See above.