The case that Joe Biden will defeat President Trump rests mainly on the polls. The case that Trump will win rests at least in part on the fact that the polls got the 2016 race wrong.
But differences in the ground rules of the two elections suggest that Hillary Clinton might have defeated Trump in 2016 under this year’s conditions.
First, consider all of the early voting this year. In 2016, it is thought that Trump made up ground on Clinton in the closing weeks and days of the campaign, especially in the key states of Pennsylvania, Michigan, and Wisconsin, all of which went to Trump by very small margins.
This year, making up ground in the final weeks and days (if Trump is able to) may be considerably less helpful than it was four years ago because a much higher percentage of the vote will be cast early. Thus, to the extent that Trump’s narrow victories four years ago in the Pennsylvania, Michigan, and Wisconsin were a function of a late surge, he may not be able to pull out wins in some or all of these key states this year even if he’s as popular on election day 2020 as on election day 2016.
Next, consider the opportunities for voting fraud this year. The 2016 election was held largely in the traditional way. This year, voting by mail is far more prevalent. This creates many more opportunities to manipulate the outcome.
If these same opportunities had existed in 2016, the Democrats might have stolen some or all of the key states for Hillary.
There is also the matter of third party candidates. Four years ago, the vote of Jill Stein, the Green Party candidate, exceeded Trump’s margin in victory in Pennsylvania, Michigan, and Wisconsin. (However, the vote of Gary Johnson, the Libertarian Party candidate exceeded Stein’s in all three states.)
This year, the Green Party apparently isn’t on the ballot in any of these three states. (The Libertarian Party is, but with a different, less prominent candidate.) The absence of a visible far left alternative might help Biden if the race is extremely close.
In any case, the other two factors — early voting and voting by mail — are potentially significant, in my view. They raise the real possibility that the same voting preference patterns displayed by the electorates of Pennsylvania, Michigan, and Wisconsin four years ago might produce Democratic victories in some or all of these states this time around.