US 2020 Presidential Election Predictions

This is the only site with all 2020 US Election Predictions

Nov 8, New York Times (Upshot) Prediction.

Sabato, Cook and Roth kept their predictions qualitative, with a "Lean Democrat" rating, not willing to get very specific. See Upshot website .

Almost all prediction models considered the Democratic candidate Hillary Clinton was most likely to win the elections. They were wrong. Donald Trump won the election, with 304 EV's, compared with Hillary Clintons 227 EV's. The above map is from the website 270towin.com which provides many historical maps and insight. The states won by Trump are colored in red. The popular vote was in Clinton's favor, approximately 65 million to 62 million.

How wrong were the prediction models? The best summary I could find of the 2016 predictions is the Upshot website, a New York Times website which was free to non-subscribers. The odds of Hillary Clinton chance of winning had risen from 60% in June 2016 to 85% just before the election. Their prediction was based heavily on the state polls.

Nate Silver's www.fivethirtyeight.com predicted a 71% chance that Hillary Clinton would win, while the Huffington Post and Daily Kos website, put the chance of a Clinton victory at 98% and 92%, respectively. So, the 538 website, although wrong, wins the award for being least wrong. as the others.

The 538 website map, as shown above from Nov 2016, called every single one of the Solid-Republican correctly. The website's predictions were right in the Republican Likely to Lean category as Trump won Alaska, Arizona, Iowa, Georgia and Ohio. But, Trump also won North Carolina, Michigan, Pennsylvania, Wisconsin, New Hampshire and Florida, all considered slightly more likely to vote Democrat judging by polling data. Every single one of the these states were won by less than a 5% margin.

The 2020 election is likely to be similar to 2016, in each candidate will focus on the "battleground states." For forecasters, they may be toss-ups but not to the candidates. Given the 2016 experience, the states which could go to either side are not just toss ups, but the tilts, leans and likelies.

There's also a strange twist we've learned form 2016, Hillary's strong and early lead in the polls, made it less likely that her supporters felt an urgent need to vote for her in November. When the Huffington Post, a generally liberal newspaper, predicted a 98% chance of Clinton winning, this might have hurt her chances of being elected. There is an underdog advantage of being behind in the polls.

Every party stategist knows where the other candidate is weak and the state is vulnerable to being flipped in 2020. Three of the four states which were won by less than 1%, were won by Trump: Michigan (16), Pennsylvania (20) and Wisconsin (10). Clinton won New Hampshire by less than one percent.

The 538 website had identified in 2016, the four top "tipping point" states: Florida, Pennsylvania, Michingan and North Carolina, which were identified as states the winning candidate must win. All four were won by Trump. In sum, 2020 is shaping up to be a lot like 2016 as these states will all be hotly contested.

I will not review the 2016 elections, and what went right or wrong. There are plenty of analyses, including a book by Hillary Clinton. The really experienced forecasters got their final prediction wrong, but I feel they were being nonpartisan and just going with the best available polling data.

The challenges in 2016 election forecasting likely will be with us in 2020. Many have pointed out that Trump was portrayed as the "new energy" but today he is the incumbent, so he may be judged differently. The Democratic contenders are well known, so voters looking for the outside candidate to shake things up, may be disappointed.

The winning formula is not so much as convincing the undecided voters to join one side, but to generate enough enthusiasm among supporters for them to make the effort to vote. The energy has to flow from their ears of the electorate to their legs. So, in 2020, how does one factor in an enthusiasm factor into the election prediction?

Included below are links to Amazon's books on the 2016 elections. Wikipedia is another great source of unbiased information.

Clinton, H. What Happened, September 2017 (Amazon.com link)

Silver, Nate, The Signal and the Noise: Why So Many Predictions Fail--but Some Don't, Reprint edition, 1995.

Lichtman, A., The Keys to the White House, 2008.

According to Amazon, there are over 2,000 books available on the 2016 election, both before and after the elections. That's a lot of reading.