Five lessons from our US presidential MRP model
Sign up here to receive our regular Election Note! Read the full analysis and receive the full version of each note as soon as it is published.
Published 15 November 2023
Key findings
Trump would narrowly beat Biden to the White House if the election were held today. If the US presidential elections were held today, Donald Trump would beat Joe Biden by 292 to 246 electoral college votes.
Biden still leads the popular vote by 49% to 48%. Compared to 2020, this lead has narrowed from 4.5 percentage points to just 1 percentage point.
The race in several battleground states is currently extremely close. Four states are currently won by a margin of less than 2% of the vote. Trump is ahead in Arizona, Georgia, Wisconsin and Pennsylvania, while Biden currently holds onto Nevada and Michigan.
Biden outperforms both Kamala Harris and Gavin Newsom, while Trump performs significantly better than DeSantis. At this point in the race, it seems like Trump and Biden are their parties’ best bets for winning the presidency.
RFK Jr.’s potential entry into the race would slightly benefit Trump. However, it is likely that he would draw voters from both major parties, making his impact on the election results relatively small.
A year out, Trump has a small lead on Biden in the race to the White House
Approximately a year before the 2024 US presidential election, this week Stack published its first set of MRP model estimates for the race. As the largest poll (n=15,205) and the only MRP at this point in the cycle, the results provide an early view of the state of the race before the start of the primaries. The results were covered at length in POLITICO and The Hill in the US, as well as in The Telegraph and The Independent in the UK. This note provides a quick overview of what we think are the most striking toplines. If you would like to explore these results in more detail, our interactive dashboard of the results can be found here.
If the US presidential elections were held today, our mean MRP model estimates show that Donald Trump would beat Joe Biden by 292 to 246 electoral college votes (see Figure 1). Specifically, this result is powered by Trump winning back four key swing states that Biden won in 2020: Arizona, Georgia, Pennsylvania, and Wisconsin. All of these states were won by Trump in the 2016 election before flipping to Biden in 2020.
Figure 1: Electoral College map of the Biden vs Trump scenario without RFK JR. and West
Given that we are seeing a general swing towards Trump compared to 2020, these flips are not surprising. These four states had the closest margins out of all states in 2020, and Biden only won Arizona, Georgia and Wisconsin by less than one percent of the vote. Our MRP model also shows Trump extending his lead in Florida (now R+7), North Carolina (R+3) and Texas (R+7). On the other hand, the Biden campaign is able to draw hope from the fact that, despite low Biden approval numbers, he would currently hold onto the swing states of Nevada (D+1) and Michigan (D+1). If we take these numbers as the starting point of the campaign, this leaves multiple paths to victory for the Democrats. Generally, they would “only” have to hold onto two of the four states that the Republicans are currently on course to flip.
2. Biden still leads the popular vote by 49% to 48%
As things currently stand, Biden would still win the popular vote, albeit by a smaller margin than in 2020. On average, we find a swing towards Trump of 3.3 percentage points in the nationwide popular vote, reducing Biden’s lead from 4.5 percentage points in 2020 to just 1.2 percentage points in our current view of 2024. Remember that, because of the Electoral College, winning the popular vote does not mean winning the White House.
Figure 2: Map of two-party swing by county
However, despite an average swing of R+3.3, there is a large amount of variation in swing across states and within states. This is illustrated in Figure 2, which maps the two-party swing to the county-level. Two key trends stand out:
There are a number of red states, where Trump significantly outperforms the average national swing towards him. This is particularly visible in the South East, where states like Alabama (R+4 compared to 2020), Florida (R+4), Georgia (R+4), and South Carolina (R+5) are moving towards Trump at a faster rate than the US average.
Democrats are able to overcome an unfavourable national environment and make gains within specific states. We find swings towards the Democrats in Montana (D+1 compared to 2020), Oklahoma (D+1), and West Virginia (D+2). The Democrats also outperform their national average in states like Nevada (R+1), New York (R+1), North Carolina (R+1), and Texas (R+1).
3. The race in several battleground states is currently extremely close
Currently, we find extremely tight races in Arizona (R+1.4), Michigan (D+0.7), Nevada (D+1.4), and Wisconsin (R+0.9). Notably, our model finds Trump underperforming his national average in all four of these states - a hopeful sign for Democrats. One possible explanation lies in the fact that these four are all swing states. As a result, campaigns in these areas are likely to start significantly earlier, as these areas are prioritised by parties. Read this way, Trump’s underperformance could well be seen as an early success by the Democratic campaign to contain Republican gains within these crucial states.
In Georgia and Pennsylvania, our model indicates that Trump currently holds a stronger lead of R+3.3 and R+2.3 respectively. Both states swing in line with the national average, turning very tight Biden wins in 2020 into Trump wins in our current projection. It is worth noting that in our second modelled scenario, including RFK Jr. and West in the race, these margins change slightly. In that case, Trump would be set to win Pennsylvania by a larger margin (R+2.7) but Georgia by a smaller margin (R+3.1). Further research is needed to establish whether or not these trends persist, but this does illustrate the complex effect that minor candidates can have in swing states.
4. Biden outperforms both Kamala Harris and Gavin Newsom, while Trump performs significantly better than DeSantis
Trump and Biden are considered to be the most likely candidates for president for their respective parties. However, both are facing challengers as well, with a number of primary competitors for Trump and calls from senior Democrats to step aside for Biden. We tested whether candidates other than Trump or Biden would perform better for their respective parties. However, our MRP model shows no advantage to either party in changing their candidate. With regard to different candidates on the Democratic side, we find that - as things stand - Donald Trump would beat both Gavin Newsom and Kamala Harris by larger margins. Specifically, Trump would currently win 311 electoral college votes against Harris’ 227. Compared against the Trump v Biden scenario, Trump would flip Michigan and New Hampshire in addition to his gains against Biden. In the scenario where Newsom would represent the Democratic Party, Trump would win by 319 electoral college votes to Newsom’s 219. Compared to the Harris scenario, Nevada would also be flipped by the Republicans.
On the Republican side, we find that Joe Biden would beat potential candidate Ron DeSantis by a score of 359 electoral college votes to 179. In this scenario, Biden would hold onto all his 2020 gains. In addition, he would pick up North Carolina and Texas to win the electoral college by a large margin. These modelled estimates are visualised in Figure 3.
Figure 3: Electoral College map of the Biden vs DeSantis scenario
5. RFK Jr.’s potential entry into the race would slightly benefit Trump
Figure 4: Voter flows between the 2020 elections and our current estimates of the 2024 elections (scenario with Trump, Biden, RFK and West)
On average, we find that RFK Jr.’s entry into the race would slightly benefit Trump. As Figure 4 illustrates, approximately 8.2% of voters who voted for Biden in 2020 would vote for RFK Jr. if elections were held today. In absolute terms, this would cost Biden approximately 5.25 million votes. Out of Trump’s 2020 electorate, we estimate that 6.9% would vote for RFK Jr. if elections were held today. This is equivalent to 4.6 million voters. Overall, from a two-party perspective, RFK Jr’s entry into the race would represent a net loss of approximately 625,000 votes for Biden based on this poll.
While RFK Jr. disproportionately hurts Biden nationwide, this is not the case in all states. In swing states Georgia and Michigan, for example, Joe Biden’s statewide margin is very slightly improved in the scenario where third parties are included (remaining at Trump+3 and Biden+1, respectively). In Pennsylvania and Wisconsin, the effect is slightly negative (going from Trump+2 to Trump+3 in PA, and Trump+1 to Trump+2 in WI), while in Arizona and Nevada third party candidates are a significant drag on the Democratic Party vote (R+1 to R+3 and D+1 to R+0, respectively). These state-level differences speak to the differing makeup of the Democrat and Republican coalitions in these states, with RFK Jr. appealing across traditional party lines.
We are also keeping an eye on:
US primaries - With another presidential election year coming up, primary season is fast approaching. Kicked off by the Iowa caucus on January 15th, the Republican primaries are particularly interesting. Will ongoing legal proceedings affect Trump’s performance? And can any other candidate position himself/herself as a viable alternative?
European Parliament elections - After a tumultuous year in European politics, citizens of all 27 EU member states will head to the polls to elect their European representatives. Current polling shows a weakening lead for the EPP and increasing strength of the flanks of the Parliament. If that turns out to be true, coalition building will become increasingly complicated in the EP’s next term. Voting 6-9 June 2024.