Vinod B
Published in

Vinod B

Final Election Forecast: What Will Happen in the President and Senate Elections on November 3, 2020

In previous posts we provided detailed senate and presidential forecasts for the upcoming elections on November 3.

Here we update those forecasts with the latest polls as of November 2, the day before election day.

Presidential Election Update

Using the latest economic data and national and state polls, we have seen the fundamental and hybrid model forecasts converge. Biden now has an 80% probability to win according to both models.

The alignment of fundamental and hybrid forecasts emphasizes Biden’s stronger position than Clinton in 2016 when the two forecasts diverged. In 2016 the fundamental forecast favored Trump while a hybrid forecast using polls favored Clinton.

As the previous in depth analysis showed, there has been a national swing towards Biden, especially in the key competitive states, that increase his odds compared to polls in 2016. Similarly, the economy has been suffering from COVID related slowdowns (despite its improving prospects since May) both nationally and locally in key states in ways that historically have forecasted Democratic victories.

While the top line forecasts have not changed very much (with a movement in the fundamentals forecast towards Biden from 72% to 80%), is there something we can learn from recent poll movements nationally and in key states?

Here is the movement in national polls over the month of October after adjusting for historical pollster accuracy and house effects (but not any time trend in polls). We see that over the last week, Biden’s national percentage has declined slightly along with Trump’s, which has declined a bit more, albeit both with fairly wide errors. This suggests that there are still some undecided voters heading into election day, highlighting why the overall forecast has not changed in the past few days.

Digging deeper to the state level, in the in depth forecast we identified Trump’s overall poor electoral prospects where he needs to hold six key states in order to win: Pennsylvania, Florida, Ohio, North Carolina, Arizona, and Iowa.

Recent polls in these states show differing movements. The race in Pennsylvania appears to have tightened with Trump gaining voters to bring the state margin to just under 5%. This is closer to the historical margin of error in state polls, suggesting that Trump’s overall chances to win the state have risen. Indeed, Biden’s probability to win the state has dropped slightly from 82% to 80% as a result of recent poll movements.

Florida and Ohio show little movement though, while recent polls would suggest Trump is winning in North Carolina and Arizona but losing in Iowa to Biden (ever so slightly). These things somewhat wash out in simulations given that Trump’s electoral math require him to hold all these states, making the top line forecast unchanged.

Recent polling results have to slightly concern the Biden camp since most states show a trend towards Trump, suggesting that he could be picking up late deciding voters. If recent poll trends continue, Pennsylvania, North Carolina, and Arizona will be more competitive than expected.

Perhaps more concerning for Biden, other Midwestern states show similar trends of movement towards Trump in very recent polls. Except for Minnesota, both Michigan and Wisconsin appear to have tightened somewhat or have an increase in the number of undecided voters. Still, while Biden’s lead has narrowed, he does hold a lead overall in these states, and he is still in a very strong overall position. For example, if Biden were to lose Michigan, Wisconsin, and Pennsylvania but other states follow expected patterns (meaning those states’ results are not due to a late breaking national swing towards Trump), Biden would still win the election with around 287 electoral votes.

While there have been movements within states and an increase in the number of undecided voters, overall the Presidential election results look quite similar to what they were just a few days ago.

Senate Election Update

The fundamental and polls hybrid model currently gives Democrats a strong chance of winning the senate with a 66% chance of winning more than 50 seats and therefore an outright majority.

There is a 10% chance that both parties will win 50 seats, leading to a tie, and a 24% chance that Republicans win more than 50 seats. If both parties gain 50 seats in the senate, ties are broken by the vice president, which is determined by who wins the presidential election. A hybrid forecast of the presidency gives Democrats an 80% chance of winning and Republicans a 20% chance. Therefore, Democrats are likely to control the senate roughly 74% of the time (66%+80%*10%). This is compared to the 71% as of October 28, meaning Democrat’s chances have increased by 3%.

In the previous in depth forecast, we identified the four closest states with a forecasted Democratic win probability between 40% and 60%: Maine, Iowa, Montana, and Georgia.

Here are recent senate polls in those states. Movement suggests democratic prospects appear to have improved in Maine and Montana, stayed roughly the same in Iowa, and worsened slightly in Georgia.

And here are forecasted probabilities in those states as of October 28 and November 1.

While polls in Georgia appear to have moved towards Republicans, that is the result of a single poll post October 28, meaning the result is uncertain. Taking an average of all recent polls, even after updating for recent trends, suggests a slight move towards the Democrats. The more favorable prospects overall for Democratic candidates in three of these four states, Sara Gideon (Maine), Steve Bullock (Montana), and Jon Ossoff (Georgia), is the reason for their increased changes to take the senate.

The other races are mostly unchanged, meaning these are still the states to watch for on election day.

A Final Call to Vote!

While polls and economic factors are predictive of elections, at the end of day turnout is what determines results. Everyone should make sure to vote on election day.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Vinod Bakthavachalam

Vinod Bakthavachalam


I am interested in politics, economics, & policy. I work as a data scientist and am passionate about using technology to solve structural economic problems.