Was this a "legitimate" election? (Part I)

This is an interesting question and the answer is not a simple yes or no.... 

When most of the country went to bed on election night, they did so believing that Donald Trump was going to win reelection. 

• Trump had won Florida in impressive fashion, beating his 2016 numbers across the board in pretty much every demographic. 
• Trump had won Ohio by the same eight points he won in 2016. 
• Trump had won in Iowa, improving slightly on his 2016 numbers. 
• NYT needle had Trump at a greater than 95% chance of winning North Carolina 
• NTY needle had Trump at approximately 85% chance of wining Georgia (which they projected he would win by over four points). Georgia election officials said they were done counting and would resume in the morning. 
• Trump was leading by nearly 700K in Pennsylvania, over 300K in Michigan, and over 100K in Wisconsin. All three states announced (almost simultaneously) that they were done counting for the night and would resume in the morning.
• The betting odds went from Biden being a 2-1 favorite to Trump being an 8-1 favorite (at around the same time that Georgia and the three great lake states announced that they had stopped counting for the night). 

Now the betting odds had been moving all night. Not just because of the states that had been called and the large leads Trump had in the great lakes battleground states. But largely because of the early bellwether precincts that were coming in across the country. Whether you were watching the 538 live update thread, following election experts on twitter, or watching cable news, they were all saying the same thing. Nearly every early precinct reporting, with every available demographic breakdown, was coming back with Trump overperforming 2016. Or if you were reading 538, they kept telling you that these early returns were a bad omen for Biden. Sure, there was the occasional bright spot for Biden. But not enough to really sway anyone. The 538 crew and liberals on twitter were already reflecting on what went wrong. 

Now these experts were smart and careful, as were the bettors. They didn't look at partial returns, knowing that some places were counting mail in ballots first and others were counting in person votes first. They were not looking at the great big leads for either candidate in states like Michigan or Arizona. They were relying on those precincts (within counties) that reported full results. More to the point, some of these bellwether precincts were rural, some suburban, some urban. Some had mostly white voters, while others were more diverse. But in pretty much every case, there was reason to believe that Trump was looking good and that Biden was in big trouble.

Meanwhile exit polling was showing that Trump had improved his standing with every single demographic except white males and that the Republican base had come out in the sort of large numbers that we had not seen since 2004. 

On top of that, Democrats were getting crushed in the down ballots. They were losing Congressional seats. Endangered Senators (like Susan Collins and Thom Tillis) were winning by comfortable margins. Expected close races (Ernst, McConnell, Graham) turned into laughers. Democrats thought they had something going in Texas but Trump was up by six and Cornyn was flirting with a double digit win. Democrats were demoralized all over the place and for good reason. 

If you wonder why the betting odds were giving Trump nearly a 90% chance at victory, it's because there was pretty much zero signs on election night (or even creeping past midnight into the next day) that this was going to be Biden's election. 

So what happened? 

Stay tuned for Part II