Overview

Welcome! This website holds our forecasts for the 2020 U.S. elections. Our work centers around compartmental modeling and uses the methods introduced by A. Volkening, Daniel F. Linder, Mason A. Porter, and Grzegorz A. Rempala in this article. It is part of our undergraduate research mentored by Alexandria Volkening at Northwestern University.

 
Our forecasts are probabilistic: we simulate 10,000 stochastic elections and calculate the mean vote margins across our simulations. To visualize uncertainty, we show a random sample of 500 simulated elections in the video on the right.
 
To reproduce or build on our forecasts, all of the model code is available here.

 

– Samuel Chian, William He, and Christopher Lee

State groupings

For the presidential elections, we forecast the vote margin in swing states individually and group the other states into either safe red or safe blue "superstates", in which we forecast the mean vote margin. Click below to expand the breakdown of our swing states and superstates.
 
State categorizations
Swing states:
  • Arizona (AZ)*
  • Colorado (CO)
  • Florida (FL)
  • Georgia (GA)*
  • Iowa (IA)
  • Michigan (MI)
  • Minnesota (MN)
  • Nevada (NV)
  • New Hampshire (NH)
  • North Carolina (NC)
  • Ohio (OH)
  • Pennsylvania (PA)
  • Texas (TX)*
  • Virginia (VA)
  • Wisconsin (WI)
We forecast the vote margin in each of these states individually.
 
*Arizona is considered a swing state in our model starting with our forecast on September 3, Texas starting with our forecast on September 27, and Georgia starting with our forecast on October 5. For our forecasts prior to these dates, the respective state was considered part of the Red superstate.
Red superstate:
  • Alabama (AL)
  • Alaska (AK)
  • Arkansas (AR)
  • Idaho (ID)
  • Indiana (IN)
  • Kansas (KS)
  • Kentucky (KY)
  • Louisiana (LA)
  • Mississippi (MS)
  • Missouri (MO)
  • Montana (MT)
  • Nebraska (NE)*
  • North Dakota (ND)
  • Oklahoma (OK)
  • South Carolina (SC)
  • South Dakota (SD)
  • Tennessee (TN)
  • Utah (UT)
  • West Virginia (WV)
  • Wyoming (WY)
We forecast the mean vote margin across these states.
 
*We consider Nebraska as one collective, rather than three congressional districts.
Blue superstate:
  • California (CA)
  • Connecticut (CT)
  • Delaware (DE)
  • Washington, DC (DC)
  • Hawaii (HI)
  • Illinois (IL)
  • Maine (ME)*
  • Maryland (MD)
  • Massachusetts (MA)
  • New Jersey (NJ)
  • New Mexico (NM)
  • New York (NY)
  • Oregon (OR)
  • Rhode Island (RI)
  • Vermont (VT)
  • Washington (WA)
We forecast the mean vote margin across these states.
 
*We consider Maine as one collective, rather than two congressional districts.
 

Final Forecast of the 2020 U.S. Presidential Election

According to our final forecast, Georgia, Iowa, Nevada, North Carolina, Ohio, and Texas have the closest races, with less than a 2.5 percentage-point margin of victory for the leading candidate. We give former Vice President Joe Biden an 88% chance of winning. Our model forecasts a victory for President Donald Trump about 11% of the time. Our forecast projects that Biden will receive an average of 350 electoral votes, and Trump will receive an average of 188 electoral votes. Importantly, these numbers are based on the assumption that undecided voters do not strongly gravitate toward one candidate or the other. Our forecasts use polling data that we downloaded from FiveThirtyEight on November 2 to determine our model parameters. Go vote, and then check back after the election for an analysis of our forecast performance!

Graph of the chance of different electoral vote outcomes
 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
 

Our forecast vote margins for each state and superstate (lines indicate the range in which the middle 80% of our simulated election results lie)

 

In our state ratings, we show the percentage of our 10,000 simulated elections in which Trump or Biden wins each state. To account for uncertainty, our simulated elections are all slightly different. For example, Biden wins in Florida in about 78.5% of our simulated elections today (down from about a 90% chance in August), and Trump wins in about 21.5% of these elections. This means we forecast that Trump has roughly the same chance of winning Florida as you do of getting two heads when you flip a coin two times (25%).


We posted our final forecast on November 2 at 21:40 CST.

 

Prior Forecasts

You can access the forecasts that we made for the 2020 presidential elections on earlier dates by clicking the links below to expand the page.
 
October 31, 2020, 22:40 CDT

We forecast that Georgia, Iowa, Nevada, North Carolina, Ohio, and Texas have less than a 2.3 percentage-point margin of victory for the leading candidate. We find that former Vice President Joe Biden's chance of winning today is about 89%. Our model forecasts a victory for President Donald Trump about 11% of the time. Importantly, these percentages are based on the assumption that undecided voters do not strongly gravitate toward one candidate or the other. Our forecasts use polling data that we downloaded from FiveThirtyEight on October 29 to determine our model parameters.

Graph of the chance of different electoral vote outcomes
 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
 

Our forecast vote margins for each state and superstate (lines indicate the range in which the middle 80% of our simulated election results lie)

 

In our state ratings, we show the percentage of our 10,000 simulated elections in which Trump or Biden wins each state. To account for uncertainty, our simulated elections are all slightly different. For example, Biden wins in Florida in about 78.3% of our simulated elections today (down from about a 90% chance in August), and Trump wins in about 21.7% of these elections. This means we forecast that Trump has roughly the same chance of winning Florida as you do of getting two heads when you flip a coin two times (25%).

October 29, 2020, 21:00 CDT

We forecast that Georgia, Iowa, Nevada, North Carolina, Ohio, and Texas have less than a 2.4 percentage-point margin of victory for the leading candidate. We find that former Vice President Joe Biden's chance of winning today is about 88%. Our model forecasts a victory for President Donald Trump about 12% of the time. Importantly, these percentages are based on the assumption that undecided voters do not strongly gravitate toward one candidate or the other. Our forecasts use polling data that we downloaded from FiveThirtyEight on October 28 to determine our model parameters.

Graph of the chance of different electoral vote outcomes
 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
 

Our forecast vote margins for each state and superstate (lines indicate the range in which the middle 80% of our simulated election results lie)

 

In our state ratings, we show the percentage of our 10,000 simulated elections in which Trump or Biden wins each state. To account for uncertainty, our simulated elections are all slightly different. For example, Biden wins in Florida in about 77.4% of our simulated elections today (down from about a 90% chance in August), and Trump wins in about 22.6% of these elections. This means we forecast that Trump has roughly the same chance of winning Florida as you do of getting two heads when you flip a coin two times (25%).

October 28, 2020, 20:23 CDT

We forecast that Georgia, Iowa, Nevada, North Carolina, Ohio, and Texas have less than a 2.5 percentage-point margin of victory for the leading candidate. We find that former Vice President Joe Biden's chance of winning today is about 89%. Our model forecasts a victory for President Donald Trump about 11% of the time. Importantly, these percentages are based on the assumption that undecided voters do not strongly gravitate toward one candidate or the other. Our forecasts use polling data that we downloaded from FiveThirtyEight on October 27 to determine our model parameters.

Graph of the chance of different electoral vote outcomes
 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
 

Our forecast vote margins for each state and superstate (lines indicate the range in which the middle 80% of our simulated election results lie)

 

In our state ratings, we show the percentage of our 10,000 simulated elections in which Trump or Biden wins each state. For example, Biden wins in Florida in about 78.6% of our simulated elections today, and Trump wins in about 21.4% of these elections. This means we forecast that Trump has roughly the same chance of winning Florida as you do of getting two heads when you flip a coin two times (25%).

October 27, 2020, 21:58 CDT

We find that former Vice President Joe Biden's chance of winning is about 88%. Our model forecasts a victory for President Donald Trump about 12% of the time. Importantly, our forecasts are based on the assumption that undecided voters do not strongly gravitate toward one candidate or the other. In Florida, we are seeing a slight decrease in Biden's chances: we give Biden about a 78% chance of winning Florida today, but we gave him about a 90% chance at the beginning of August, and about an 85% chance at the start of October. Our forecasts use polling data that we downloaded from FiveThirtyEight on October 25 to determine our model parameters.

Graph of the chance of different electoral vote outcomes
 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
 

Our forecast vote margins for each state and superstate (lines indicate the range in which the middle 80% of our simulated election results lie)

 

Georgia, Iowa, Nevada, North Carolina, Ohio, and Texas have less than a 3% margin of victory for the leading candidate, indicating toss-up races. In ten states, the percentage of voters that remain undecided at the end of our simulated elections is higher than the vote margin that we are forecasting for the leading candidate. If undecided voters turn out heavily in favor of one candidate or the other, we could easily see these states changing from blue to red, or vice versa.

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
October 25, 2020, 23:46 CDT

Using polling data from FiveThirtyEight that we downloaded on October 23, we find that former Vice President Joe Biden's chance of winning is holding steady at about 89%. Our model forecasts a victory for President Donald Trump about 10% of the time. Many states have close races, and, in six states (Georgia, Iowa, Nevada, North Carolina, Ohio, and Texas), we are foreasting less than a 3% margin of victory for the leading candidate.

Graph of the chance of different electoral vote outcomes
 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
 

Our forecast vote margins for each state and superstate (lines indicate the range in which the middle 80% of our simulated election results lie)

October 23, 2020, 23:26 CDT

Using polling data from FiveThirtyEight that we downloaded on October 21, 2020, we find that former Vice President Joe Biden's chance of winning is holding pretty steady at about 89%. Our model forecasts a victory for President Donald Trump about 11% of the time. Importantly, there are many states with close races. In six states (Georgia, Iowa, Nevada, North Carolina, Ohio, and Texas), we are foreasting less than a 3% margin of victory for the leading candidate. This is roughly the margin of error in many polls.

Graph of the chance of different electoral vote outcomes
 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
 

Our forecast vote margins for each state and superstate (lines indicate the range in which the middle 80% of our simulated election results lie)

October 19, 2020, 16:45 CDT

Using polling data from FiveThirtyEight that we gathered on October 15, 2020, at 22:22 CDT, our model forecasts a victory for Joe Biden 87.80% of the time, and a victory for Donald Trump 12.01% of the time.


In this forecast, we see the Red superstate move more Republican, and we give Trump slightly higher odds of winning than in the last forecast, but similar to forecasts before that.

Graph of the chance of different electoral vote outcomes
 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
 

Our forecast vote margins for each state and superstate (lines indicate our 80% confidence intervals)

October 5, 2020, 22:43 CDT

Using polling data from FiveThirtyEight that we gathered on September 28, 2020, at 20:21 CDT (just prior to the first presidential debate), our model forecasts a victory for Joe Biden 90.99% of the time, and a victory for Donald Trump 8.87% of the time.


As an adjustment to our model, we now forecast Georgia individually as a swing state instead of as part of the Red superstate. We currently forecast Georgia as leaning slightly Democratic. In this forecast, the Red superstate also has a larger forecast Republican margin than before, and it rarely votes Democratic in our simulated elections now. This is what we expect of the superstate, as the states within it are more solidly Republican than Georgia according to the polling data.

Graph of the chance of different electoral vote outcomes
 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
 

Our forecast vote margins for each state and superstate (lines indicate our 80% confidence intervals)

September 27, 2020, 23:30 CDT

Using polling data from FiveThirtyEight that we gathered on September 21, 2020, at 19:20 CDT, our model forecasts a victory for Joe Biden 87.86% of the time, and a victory for Donald Trump 11.86% of the time.


We have made an adjustment to our model: we now forecast Texas individually as a swing state instead of as part of the Red superstate, and will continue doing so going forward. This change reflects the uncertainty surrounding the vote in Texas this year, and our model currently forecasts it as leaning slightly Democratic. By bringing Texas out of the Red superstate, the Red superstate now contains states that are more safely Republican, which may have contributed to it being forecast slightly more Republican. It still went Democratic in just under 5% of our simulations, but this is less than before.

Graph of the chance of different electoral vote outcomes
 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

Bar graph of the forecasted vote margin and 80% confidence intervals for each state
 

Our forecast vote margins for each state and superstate (lines indicate our 80% confidence intervals)

September 14, 2020, 20:00 CDT

Using polling data from FiveThirtyEight that we gathered on September 13, 2020, at 22:00 CDT, our model forecasts a victory for Joe Biden 86.74% of the time, and a victory for Donald Trump 12.82% of the time.


In this second forecast with Arizona considered a swing state, our projections did not change significantly from our prior forecast 10 days ago. Overall, Trump's odds of winning increased very slightly.

 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

 

Our forecast vote margins for each state and superstate (lines indicate our 80% confidence intervals)

September 3, 2020, 23:00 CDT

Using polling data from FiveThirtyEight that we gathered on September 3, 2020, at 23:00 CDT, our model forecasts a victory for Joe Biden 87.26% of the time, and a victory for Donald Trump 12.42% of the time. This forecast reflects new polls that were conducted following the Democrat and Republican National Conventions.


Additionally, we made an important adjustment to our model: we now forecast Arizona individually as a swing state instead of as part of the Red superstate, and will continue doing so going forward. We believe this change enables our forecasts to better reflect the uncertain nature of Arizona this year (our model currently forecasts it going Democrat), as well as ensure that the Red superstate is properly composed of safe red states that will not skew its forecasted vote margin towards Democrat.


This recategorization of Arizona may have helped the Red superstate be more solidly Republican, and its 80% error bar no longer crosses over to Democrat. Still, it goes Democrat in over 5% of our simulations (although less often than before). Other changes in our forecasts include Iowa and Ohio moving more Republican, and our electoral vote distribution becoming more spread out across multiple possible outcomes, with less of a concentrated spike at the point representing all swing states going Democrat. Overall, the election odds for each candidate are similar to what they were on the previous forecast; Arizona being split off as a swing state and forecasted as leaning Democrat may have balanced out with Red superstate, Iowa, and Ohio moving Republican.

 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

 

Our forecast vote margins for each state and superstate (lines indicate our 80% confidence intervals)

August 14, 2020, 23:00 CDT

Using polling data from FiveThirtyEight that we gathered up until August 14, 2020, at 23:00 CDT, our model forecasts a victory for Joe Biden 87.17% of the time, and a victory for Donald Trump 12.7% of the time.


Notable changes from our prior forecast include that the red superstate now has a greater forecast margin of Republican victory; this is the main contributor to the slightly greater odds that we now give Donald Trump. Meanwhile, Ohio has flipped from slightly Republican to just barely Democrat, with a forecasted margin of victory near zero. Iowa and North Carolina did not flip to the other party, but the forecasted margins of both are now closer to zero, indicating very close races. Our forecasts for other states stayed mostly the same, without significantly impacting the overall electoral vote distribution.

 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model

 

Our forecast vote margins for each state and superstate (lines indicate our 80% confidence intervals)

August 3, 2020, 19:30 CDT
 

Our forecast vote margins for each state and superstate (lines indicate our 80% confidence intervals)

 

The distribution of our simulated elections that resulted in various electoral-vote outcomes, across 10,000 stochastic simulations of our model


Here are our forecasts based on data from FiveThirtyEight that we gathered up until August 3, 2020 at 19:30 CDT.


Our model forecasts a victory for Joe Biden 88.21% of the time and a victory for Donald Trump 11.73% of the time.


One interesting point is that, in about 10% of the 10,000 elections that we simulate, the red superstate actually votes for Joe Biden. As a result, our model forecasts that there is about a 10% chance that Donald Trump will earn no electoral votes, which is clearly unrealistic. It should be noted, however, that even if we specify that the red superstate always votes Republican, those 10% of outcomes would still correspond to an overall Democrat victory (due to the swing states going Democrat), so this pheneomenon has little effect on the chances that we give each candidate of winning the election. The cause of the red superstate not being solidly Republican could be due to some states in our red superstate leaning more Democrat this year; for example, many other forecasters (such as Sabato's Crystal Ball) are currently calling Arizona a swing state.

We may want to consider reducing the number of states that we include in our superstates this year, recategorizing Arizona, for example, as a swing state. However, there are some drawbacks to this; for instance, our forecasts take longer to produce when we add more individual states into our modeling framework. Moreover, adjusting our superstate categorizations may make it more difficult to compare our 2020 forecasts to the model's performance on forecasts of past elections, which were made with the same arrangement of states in the superstates that we are using right now.

 

Our state ratings as Solid, Likely, Lean, and Toss-Up