Election Projection Analysis Summary
Without utilizing any polling data in my election projection, I predicted 48 out of 50 states correctly. I only missed Arizona and Georgia. If this was NCAA Bracketology I would have cashed in. My projection included the percentage of the vote for the winning candidate, and in some cases my estimates were remarkably close to the actual outcome. This election projection analysis reviews these percentage of vote outcomes.
In the following swing-states my projected percentage of the vote for the victor was within 1% of the actual outcome: Arizona; Georgia; Michigan; Minnesota; North Carolina. My projection was within 2% of the winning candidate’s percentage of the vote in these swing states: Nevada; Pennsylvania; Wisconsin. What follows is a deeper dive into this analysis.
It Really Was A National Landslide
Biden beat Trump by more than seven million votes! This is the second largest popular vote differentiation since the millennium. Needless to say, when six million people say “I like you better than him” that’s very convincing.
Biden beat Trump by more than 4%. Based upon percentage of the vote, this is the second largest victory by a candidate since the millennium. Biden also cleared 50% of the vote, which is something Trump never did.
The Rust Belt (aka Blue Tinted Wall)
In Pennsylvania, Biden beat Trump by more than 80,000 votes. This is almost double the number of votes Trump beat Clinton by in 2016. This represents a 120,000-vote swing (approximately a 1.75% Democratic gain). These results were likely influenced by Biden’s Northeastern Pennsylvania roots. Even factoring this in, 2020 shows Pennsylvania to be a state Democrats should win.
The results in Michigan were even more stark. Biden beat Trump there by over 150,000 votes. In 2016, Trump beat Clinton there by a mere 10,000 votes. This is a massive change in just four years. Despite Trump’s America First economic message, the home of America’s automotive manufacturing industry resoundingly voted for his opponent.
Biden took Clinton’s 2016 1.5% victory in Minnesota and turned it into a run-away 7% victory in 2020. He even outpaced Minnesota’s incumbent Senator, Tina Smith, who was also on the ballot (52.6% to 48.8%)
But Wisconsin remains the wildcard. Trump won it by the same margin in 2016 that he lost it by in 2020 (approximately 20,000 votes). Looking to 2024 the key variables remain the same: turnout among black voters; turnout among non-college educated white voters; the state’s southeast suburbia.
This election projection analysis shows that I correctly predicted Biden to win all four of these states. My Biden projection was within .5% in Minnesota, 1% in Michigan, and 2% in Wisconsin and Pennsylvania.
The South’s Election Projection
North Carolina was a nail-biter…but not really. Trump won by about 75,000 votes. To what degree did North Carolina’s extensive voter suppression efforts impact this result? How many votes did Trump’s last-minute barnstorming garner? Would the results have been different if the Democratic candidate for Senate was Jaime Harrison instead of Cal Cunningham? Despite exit polling and post-election analysis one never knows. Perhaps Stacy Abrams created a blueprint in neighboring Georgia that could be replicated here. I correctly projected Trump to win North Carolina with 49.5% – he won by 50.1%.
Speaking of Stacy Abrams, she can take a national nod for spearheading Biden’s Georgia victory. Hers was not a singlehanded approach – she built an electoral army. Her team overcame historical forces – Biden was only the second Democrat to win in Georgia since favorite-son Jimmy Carter. They also overcame the modern Republican apparatus – voter suppression endeavors taken from the pages of Jim Crow. Her efforts were supported by the increasingly educated, and increasingly Democratic, metro Atlanta region. My election projection correctly predicted that Trump would win with 50% of the vote was extremely precise (he earned 49.5%) but in the final analysis was wrong (he lost by 0.2%).
Georgia was a Deep South outlier: Trump’s total percentage of the vote decreased by 1.5% from 2016. Throughout the rest of this region, which he won in non-competitive races, his totals remained remarkably stagnant. His 2020 percent totals were within .5% in Alabama, Mississippi, and Louisiana. This just points to Georgia’s incredible uniqueness.
Florida (which is technically the south…but not really)
Technically Florida is part of the south. But it’s the truest of American melting pots with an amalgamation of residents who have migrated from northern enclaves, residents who have immigrated from southern nations, and, lest we forget, locals.
The election in Florida was supposed to be close. It was supposed to be one of the states which tilted the electoral college to the victorious candidate. It was neither. While I correctly predicted Trump to win, I undercut his victory by nearly 3%. He won by over 350,000 votes! That means Trump’s margin of victory was larger than his entire vote total in four other states which he prevailed in: North Dakota; South Dakota; Wyoming; Idaho.
The difference was his unchecked titling Biden as a socialist. As has been reported, this hit home with South Florida’s Venezuelan voters, among others, who came to America as a safe harbor from a corrupt, failing, socialist government. The twisted irony is that many of these first and second generation immigrants who voted for Trump would not have been welcomed into America under Trump’s own immigration policy.
Texas’ Election Projection
There were many, including Beto O’Rourke and Alexandria Ocasio-Cortez, who believed this was the year Texas would turn blue. They believed that high turnout and an increasingly Latin voter base, combined with Trump’s anti-Mexican immigrant message, the Democrats would prevail. The hope was that this would be the start of a blue, Texas-sized, crescendo. While this all sounded good on paper, I didn’t believe it. The hill was just to tall, as the Republican candidates have won this state for several decades – and it usually hasn’t even been close. My election projection correctly predicted Trump to win with 51.5% of the vote (he garnered 52.1%).
The West
My election projection correctly predicted Biden the winner in New Mexico and Colorado, though he easily surpassed my projection in both states. Arizona is another story. It has been viewed as a swing state for numerous elections, yet has remained a Republican stronghold. Republicans have won all but one presidential election there since 1952.
I correctly predicted Trump’s percent of the vote (I predicted 49.0% – he received 49.1%; for the purpose of this election projection analysis I would consider that within the margin of error) but I incorrectly thought he would win. He lost by 0.3% – a margin thinner than a cactus needle.
There were events which foreshadowed this result. Social worker and lawyer Kyrsten Sinema (D) won the 2018 Senate race. In 2016, Trump was the first Republican presidential candidate in recent memory who did not cross the 50% threshold. While Trump did not cross 50% of the vote, he did cross the McCain family. Repeatedly. And disgustingly. It is unknown the influence this had on the electorate, but common sense tells us this was a factor.
Also of note is that in 2020 Biden won 4.3% more of Arizona’s vote than Clinton did in 2016. While the Democratic vote increased, the Libertarian vote decreased. In 2016, the Libertarian candidate received 4.1% but in 2020 the Libertarian candidate only received 1.5%. It is plausible that Republican-leaning voters who did not support Trump voted Libertarian in 2016 and Democratic in 2020. With a more palatable Republican candidate in 2024 these right-leaning voters who vacillated between voting Libertarian (2016) and Democratic (2020) could return to the Republican fold. Any, or all, of these factors could have been the difference in Biden’s victory.