FiveThirtyEight has released its final forecast for the president. Not much has changed in the past 24 or 48 hours as most of the late polls were either close to our previous poll averages or, quite frankly, were from fairly random respondents who didn’t get much of the weight in our forecast.
Of course, you can click on the forecast now if you want to see what’s in it. I am sure most of you have already done this. But in these accompanying notes I would like to give some context. For example, when I was writing about our final 2012 prediction for president, I was trying to explain why a race that everyone assumed was close actually reflected a pretty significant advantage for Barack Obama. Conversely, when I wrote about our final forecast for 2016, it was pretty much the opposite. I tried to explain that while Hillary Clinton was favored, what most of the media portrayed safe was a highly competitive competition between her and Donald Trump.
This year … I’m not sure what to convince you of. If you think the survey was irrevocably suspended because of 2016, that is not really true. On the flip side, if it weren’t for 2016, people could look at Joe Biden’s huge lead in national polls – the largest of all candidates on the eve of the election since Bill Clinton in 1996 – and conclude that Trump certainly does will be a one year president. If that’s what you believe, please read my story earlier this week about how Trump can win and why a 10 percent chance needs to be taken seriously.
Even so, Biden’s standing at the end of the 2016 race is considerably stronger than Clinton’s. His lead is greater than Clinton’s in any battlefield state and more than twice that of the national level. Our model predicts that Biden will win the referendum by 8 percentage points, more than double Clinton’s forecast margin in late 2016.
What the voting card will look like if there is a voting error in size 2016
Indeed, some of the dynamic that allowed Trump to take off in 2016 doesn’t seem to exist this year. There are significantly fewer undecided voters in this race – just 4.8 percent of voters say they are undecided or vote for third-party candidates, compared with 12.5 percent at the end of 2016. And the polls were much more stable this year than four years ago. Finally, unlike the “Comey letter” in the last days of the campaign four years ago – when then FBI director James Comey told Congress that new evidence had emerged into the investigation of the private email server, the Clinton as Secretary of State – There has been no major development in the past 10 days to further disrupt the race.
Now there are also some sources of error that were not so relevant four years ago. The big increase in early and postal voting 100 million people have already voted! – could, for example, pose challenges to pollsters. Despite what we believe to be fairly conservative assumptions, Biden has an 89 percent chance of winning the electoral college compared to a 10 percent chance for Trump. (The remaining 1 percent reflects the rounding error and the likelihood of a tie in the electoral college.)
The tricky thing about this race, however, is that due to the benefit of Trump’s electoral college, which it largely carries over from 2016, it wouldn’t require as much polling error in Trump’s favor to make the election Interesting. Important is, Interesting is not the same as a likely Trump victory; Instead, the likely outcome of a 2016-style voting mistake would be a Biden victory, which would take time to resolve and which could jeopardize the Democrats’ chances of taking over the Senate. On the other hand, it wouldn’t require a major query error Bidens Favor to turn 2020 into a historic landslide against Trump.
So, like four years ago, let’s do some stress tests here. On average in the past elections, the final polls were around 3 percentage points behind. How would the map change if there was a 3 point mistake in Trump’s direction? And what about a 3 point mistake in Biden’s direction? Given that some states move more than others according to national trends, our final forecast shows that:
|with 3-point national error …|
|Status||538 final forecast||IN BIDEN’S FAVORITE||IN TRUMPS FAVORITE|
Before we get to the Biden or Trump friendly scenarios, let’s say this is one of those happy years where there aren’t any systematically Mistakes in the polls – that is, Biden wins nationally with about 8 points. In this case, Biden will win the electoral college, even if there may be electoral errors in individual states. Biden’s easiest route to victory would be to win back three of the so-called “blue wall” states that Hillary Clinton lost: Michigan, Wisconsin and Pennsylvania. Along with the states Clinton won in 2016, that would get Biden up to 278 votes, more than the 270 required. Pennsylvania is the Blue Wall Group’s weakest country, but even if Biden lost it – unlikely if the polls are roughly correct – he would have many other options as Arizona, Florida, in our final forecast, was also close is in front. North Carolina and Georgia, and just behind Trump in Ohio, Texas and Iowa.
What if there was a 3 point voting error in Biden’s favor? Then he would be a favorite in all of the above states. In conjunction with the 2nd Congressional Districts in Maine and Nebraska, where he is also preferred, this would result in his win 413 votes. Other states traditionally extremely red could even come into play for Biden, with Montana being the most likely option, followed by South Carolina, Alaska, and Missouri. That scenario would also result in an 11-point vote for Biden, the largest of any nominee since Ronald Reagan in 1984, and the largest margin of opportunity against an incumbent since Franklin Delano Roosevelt versus Herbert Hoover in 1932.
But with a 3-point mistake in Trump’s direction – more or less like 2016 – the race would become competitive. Biden would likely hold out, but he would only be the all-time favorite in states (and congressional districts) with 279 votes. In Pennsylvania, the turning point state, it is projected to gain 1.7 percentage points – not within the recount range, but in a tight race.
Such a scenario would not be the end of the world for Biden. The extra pillow he has relative to Clinton helps a lot; This means that with a 2016-style poll mistake, he would narrowly win some states that she narrowly lost. Biden recently did good polls in Michigan and Wisconsin and has big leads there. Still, that would not be the result the Democrats had hoped for. For one, the election could take longer, as in that scenario Biden would likely have to rely on Pennsylvania – a state that is expected to take some time to count. For another, it could make a pretty bad map of the Democratic Senate hopes, given that Biden would be a close underdog in several states with major Senate races, including Arizona, North Carolina, Georgia, and Iowa. So during Biden is not a normal sized query error removed from to loseHe’s a normal sized poll bug if he doesn’t make a messy profit that may not come with control of Congress.
As much as we tried to be careful, the Democrats have a right to be delighted with where they ended up. Sure, with a bigger electoral lead, Biden could be in a significantly safer position in Pennsylvania or Arizona, where his numbers have declined somewhat along the way. If we had told our Democratic readers six months ago that Biden would be 8 points ahead of election morning nationally, 8 points ahead in Wisconsin and Michigan, 5 points in Pennsylvania, 2 or 3 points ahead in Florida and Arizona, and even a bit ahead in Georgia and with a pretty good chance of Texas winning, we think they’d be pretty happy.
Also note the background conditions in the country today. Trump barely won the election four years ago against a highly unpopular opponent in Clinton. In 2016, 18 percent of voters took part in the national Exit poll both didn’t like Trump and Clintonand those voters voted for Trump by 17 points. If only they were evenly divided, Clinton would have (narrowly) won the electoral college. Many of these voters, in fact to like Biden, however, who has much better favoritism ratings than either Clinton or Trump.
In the meantime, the election will take place at a time when the majority of voters are 2: 1 dissatisfied with the direction of the country amid a COVID-19 pandemic that has killed – and has – 233,000 Americans got worse in the past few weeks – along with high (albeit improving) unemployment, a summer of racist protests and continual erosion of democratic norms by Trump and his administration. Trump’s approval rating has been negative for virtually all of his presidency. Trump’s election record is hardly flawless: The Democrats won the referendum for the US House in 2018 with almost 9 points, about as much as Trump now in national polls in an election where polls and forecasts were very accurate.
In other words, given what is happening in the country and Biden’s popularity compared to Clinton, it just shouldn’t be that hard to imagine a small number of voters moving from Trump to Biden. Indeed, Polls show that: There are more Trump-to-Biden voters than Clinton-to-Trump voters. The lion’s share of the people who voted for Gary Johnson or any other third-party candidate four years ago also plan to vote for Biden.
Trump card Power be able to overcome this with a disproportionately high Republican turnout. But during Republican turnout Power be very high, democratic turnout Pretty sure will also be, as evidenced among other things: Equal or higher enthusiasm among Democrats in polls;; her very high numbers of early and postal votes, and your greater fundraising skills throughout the cycle.
Again, this should not deny that Trump will also produce his voters. Our model predicts a total participation of 158 million in the race with an 80th percentile between 147 and 168 million. But when strong voters and independents mostly switch to the other party, your turnout must be high and that the other party is low to have a good shot, and that latter condition doesn’t seem likely for Trump.
Even so, there is a 10 percent chance that there has never been a choice like this, and this is not a moment anyone should take anything for granted. We hope that you will keep following our coverage until it is determined who has won.