2020 Election Predictions and Analysis

Entering 2020 the only given about this election was that there would be chaos. Few could have predicted the Coronavirus, wild economic swings, and massive scandals on both sides. The race between President Trump and former Vice President Joe Biden is finally coming to a close and I am here to make my prediction.

First let's take a look at some key metrics starting with voter enthusiasm. Trump voter enthusiasm is marginally improved from 2016 meanwhile Dem enthusiasm has almost surpassed Trump going up 25 percent. This should negate the turnout advantage Republicans experienced in 2016.

The tough part for Trump is to continue narrow wins in states where the Republican margin of error is remarkably low, looking at 2018 we saw a blue wave sweep across the nation most importantly across the Midwest where Wisconsin, Michigan, and Minnesota all saw Democratic Governors outperform polling to take significant wins. The ability to flip states that voted Clinton with Trump only having an outside chance to take Nevada and its 6 electoral votes and a long shot at Minnesota with its 10. If we grant Biden Michigan and Wisconsin of which he has leads of plus that of Hillary in 16 the electoral map tightens to needing to win one of Georgia, Florida, Arizona, Pennsylvania, North Carolina, or Ohio. All states in which you could spot Trump 2 percentage points and the results would still all be within margin of error on polling. This ultimately is crucial difference from 2016 where late polling showed a Trump surge, but not soon enough for averages to lean Trump across the board. Comparing those averages paints a dire picture for Republicans.

Across the board with the exception of Minnesota, Biden is outperforming Hillary’s numbers. This after many polls did significant adjustments to methodology to accurately project their numbers. If we remove Trafalgar Group from these averages the ratios skew heavily towards Biden. Of which there is a great case to be made for the one group whose polling predicted 2016 correctly that their methodology has skewed so far down the path of bad statistics . The cross tabs in their polls for 2020 are absolutely insane and should be readily dismissed.

Then for everyone who says “but the silent Trump voter”, this is where polling gets messy and could perhaps always be wrong. It is true many supporters of Trump shy away from telling their friends and family, important numbers in this group in 2016 where White college educated men, and conservative millennials who all greatly outperformed expected poll results. However, when one says greatly we are only talking 2–3 percentage points. This time around there are two noticeable differences, first it’s hard to find a silent Trump supporter as the rallies and social media camaraderie has largely outed most supporters. Second, both of those groups have skewed even further to the Democrats advantage this time as Trump has lost his lead amongst the key demographic for Trump, white college educated voters.

The last and final nail in the coffin to the Trump reelection odds lands on millennials and the youth vote. Already we have seen early voting numbers that have rivaled total votes cast.

This youth surge in voting at levels not seen in modern political history will greatly over perform polling models that have the youth vote participating at normal levels. A simple 3 percent increase in millennial turnout would have led to a decisive Clinton victory in 16 with wins in Pennsylvania, Michigan, Wisconsin, and Florida. Today that increase could mean wins in Georgia, North Carolina, Arizona and even Texas. Quite simply at the rate Gen X, Gen Z, and Millennials are showing up a Trump reelection is a longshot.

And thus I reach my prediction using a model built on polling averages, removing Trafalgar from the averages and assuming a slight under polling for Trump amongst Latinos and non college educated white voters of 2 percent, while factoring in an under polling of millennials by 4 percent. I reach a resounding 351–187 electoral landslide.

A few stipulations and predictions for state results. This model significantly plays on polling accurately suggesting voter turnout and that mail in ballots are counted appropriately. Given the rate of large court rulings many states could see significantly different results in large amounts of ballots are thrown out. Second Trump could still win with a few November surprises, first with a historic amount of newly registered voters for both sides in states like Florida, Arizona, Georgia, and Texas it is perfectly possible that the election turns into a race to see which sides new voters show up. The second key to a Trump victory lies in the surge of Latino support for him, something most polling fails to address by not offering polling in multiple languages. While college educated Latinos favor Biden heavily, it would be wise to not ignore the significant investment and inroads the Trump campaign has made in Texas, Florida and Arizona with Hispanic outreach and filling the airwaves of Hispanic radio stations. The last potential outcome is that voters stay home due to intimidation, corona fears, or simple political burnout. This election is unprecedented in hostility, perceived or actual risk regarding the corona virus and the duration of the campaign which seemingly began as soon as President Trump was sworn in. Either way this election could have some major surprises but I am locking in my prediction of 351–187 for the former Vice President.