PRESIDENTIAL ELECTION 2024

An analysis by TRISTIN HURST

THE KEYS TO THE WHITE HOUSE

To predict presidential elections, I adapt the methodology of Allan Lichtman. The general idea is that there are thirteen factors (true/false questions) that determine the winner of presidential elections, regardless of polling data. If six or more factors are counted against the incumbent party (in this case, the Republicans), the party will lose the next presidential election. See “https://www.socialstudies.org/system/files/2024-02/se-8801006.pdf” for a more detailed explanation.

Through this model, I attempt to translate the keys into a specific electoral college, and popular vote total.

library(mapdata)
## Loading required package: maps
library(maps)
library(scales)
## [1] "President Joe Biden (D)"
## [1] "Former President Donald Trump (R)"
## [1] "The Keys to the White House"
## 
##  Key 1. #After the midterm elections, the incumbent party holds more seats in the U.S. House of Representatives than it did after the previous midterm elections. FALSE 
##  Key 2. #There is no serious contest for the incumbent party nomination. TRUE 
##  Key 3. #The incumbent-party candidate is the sitting president. TRUE 
##  Key 4. #There is no significant third party or independent campaign. FALSE 
##  Key 5. #The economy is not in recession during the election campaign. TRUE 
##  Key 6. #Real per-capita economic growth during the term equals or exceeds mean growth during the previous two terms. TRUE 
##  Key 7. #The incumbent administration effects major changes in national policy. TRUE 
##  Key 8. #There is no sustained social unrest during the term. TRUE 
##  Key 9. #The incumbent administration is untainted by major scandal. TRUE 
##  Key 10. #The incumbent suffers no major failure in foreign or military affairs. FALSE 
##  Key 11. #The incumbent administration achieves a major success in foreign or military affairs. FALSE 
##  Key 12. #The incumbent candidate is charismatic or a national hero. FALSE 
##  Key 13. #The challenging party candidate is not charismatic or a national hero. TRUE 
##  - 
##  The Democrats hold 8 keys, the exact number needed to secure re-election.

This model already assumes that Joe Biden will win re-election, based on the methodology of Allan Lichtman. The primary objective of my analysis is to predict the electoral college result with at least 97% accuracy, and the popular vote within three percentage points.

Using election data from 2000-2020, I created a model that estimates the electoral college vote share a president can expect based on the number of true keys.

## [1] "Variables Considered in Equation"
## 
##  180 safe_votes: The number of votes that the previous Republican candidate (in 2012) won by a margin of at least fifteen percent. 
##  8.667778 change: An estimate of how many electoral votes a key is worth. 
##  6 incumbency: Incumbency equals 6 when the incumbent party president is running for re-election. 
##  8 true_keys: The amount of keys declared true by Allan Lichtman's model
## 
## 
##  EQUATION USED TO PREDICT VOTE SHARE: 
##  Predicted Result = safe_votes+((change+incumbency)*true_keys))
## 
## 
##  PREDICTED RESULT 
##  President Joe Biden (D) : 297 electoral votes 
##  Former President Donald Trump (R) : 241 electoral votes

The model now assumes that Biden will win roughly 297 electoral votes. Using the data below, we attempt to estimate where a state would fall based on its electoral trends in various elections from 2016-2022, and its estimated approval of the incumbent president.

The model takes into account presidential election results from 2016, and 2020, as well as House election results from 2022. We use morning consult polling to estimate a president’s approval in each state.

The shift variable listed below is the difference between 2018 house results, and 2012 presidential election results.

Approval ratings are the only opinion polling-based data used in this forecast. Generally, polling data has proven unreliable in prediction election results, underestimating Trump in 2016 and 2020, and overestimating Senator Mitt Romney in 2012.

Positive variables favor the incumbent party, while negative variables favor the opposing party. This is true for all numeric variables in this forecast.

Based on the data above, I use this equation: election.2024\(current_lean = ((0.5*election.2024\)shift+election.2024\(four.prez)+election.2024\)approval)/2 to estimate a state’s partisan lean relative to the rest of the nation (table below). However, this is not a prediction, as the states will shift based on the amount of true keys.

## Warning: package 'ggplot2' was built under R version 4.3.2

Below is a map of each state’s partisan lean, relative to the nation. Again, the partisan lean is not a prediction. Think of this as a battleground map, with lean states in play for either party depending upon the state of the keys.

Dark Red = Republicans +15 Medium Red = Republicans +6 Lean Red = Republicans +0 Lean Blue = Democrats +0 Medium Blue: Democrats +6 Dark Blue = Democrats + 15 

## Warning: The `<scale>` argument of `guides()` cannot be `FALSE`. Use "none" instead as
## of ggplot2 3.3.4.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

Below is our election simulator, that will run our calculations for each state, subtracting states (in order of projected margins) from the overall vote share until Trump’s electoral margin matches the one set forth by our model.

NOTE: - Five points will be added to states where a presidential candidate has won a statewide office (governor, senator, etc.) previously. - Three points will be added to states where a vice presidential candidate has won a statewide office (governor, senator, etc.) previously.

## [1] "Simulation 1"
## [1] 199
## [1] "Simulation 2"
## [1] 199
## [1] "Simulation 3"
## [1] 212
## [1] "Simulation 4"
## [1] 212
## [1] "Simulation 5"
## [1] 215
## [1] "Simulation 6"
## [1] 225
## [1] "Simulation 7"
## [1] 240
## [1] "Simulation 8"
## [1] 240
## [1] "Simulation 9"
## [1] 275
## [1] "Simulation 10"
## [1] 275
## [1] "Simulation 11"
## [1] 291
## [1] "Simulation 11"
## [1] 297
## SUMMARY OF FORECAST: 
##  President Joe Biden (D) will gain roughly 297 electoral votes in the electoral college. Former President Donald Trump (R) will gain roughly 241 in the electoral college.
## 
##  President Joe Biden (D) will win about 52% in the popular vote. Former President Donald Trump (R) will win about 46% in the popular vote.
## END OF FORECAST