Gallup asked Americans just before that election whether they thought Democrats or Republicans would control the US House after the election. By a 50% to 44% margin, they said Republicans. This came even as Democrats were clear favorites in pretty much every forecast and when that same Gallup poll showed Democrats with an 11-point lead on the generic congressional ballot. This... marked the first time that Americans incorrectly forecasted who would win the House.This leads Enten to wonder whether voters are overreacting to the 2016 election. Hillary held a small lead in the polls (which she maintained in the election contrary to much conventional wisdom), and most Americans thought she would win. She didn't, of course, and now many voters seem to believe that there's something magical about Trump, that the polls underestimate Trump's appeal. Enten disagrees:
My advice would be that in gauging the electorate, you shouldn't be of the mindset that Trump is going to pull it out if the polls continue to suggest he won't. Even Trump's own polling reportedly has him behind. Trump's a politician, not a magician.Something similar appears to be occurring with prediction markets. When asked to choose between a generic Democratic or Republican candidate, the markets currently favor the former over the latter by about 4 points (53-49). Moreover, when betting on which candidate will win individual states, the markets predict that the Democratic candidate will walk away with 290 electoral votes, while the Republican candidate will walk away with only 248. In other words, both predict a win for the Democrats. However, when asked to specifically bet on Biden or Trump, the markets choose Trump over Biden by about 6 points (50-44). Although part of this can be explained away because neither candidate has officially been chosen as their party's nominee, that's clearly not the whole story. Some of it has to reflect the sentiment that Trump's unbeatable.
I don't think he is. After 2016 the markets have become gun shy. After incorrectly predicting a Clinton win in 2016, they're overcompensating and overestimating Trump's chances of winning in 2020. Trump may win in November, but in order to do so, a higher percentage of Americans will have to vote for him then than many will now.
Note: This morning FiveThirtyEight published a "chat" ("If Trump Is Down In The Polls, Why Do So Many Americans Think He’ll Win?") that discusses this same phenomenon.