Who is leading the polls in the USA? Final update for August
A review of what the polls tell us at the end of the month.
To analyze what the polls are telling us as August comes to an end, and with only two months remaining until the elections, I recommend you take a look at the article we wrote 15 days ago, where we analyzed what the mid-month polls were indicating.
Let's get straight to the point.
In mid-August, we discussed the importance of the concepts of "swing states" and "safe states," as well as what the polls were indicating about these. Let's review those results:
Arizona: Harris 45% - Trump 45%
Georgia: Harris 47% - Trump 47%
Michigan: Harris 47% - Trump 45%
Nevada: Harris 45% - Trump 45%
North Carolina: Harris 45% - Trump 47%
Pennsylvania: Harris 48% - Trump 46%
Wisconsin: Harris 49% - Trump 46%
These data, when combined with the results from other "safe states," reflected a very narrow Democratic electoral victory, projecting 47% of the popular vote and 270 electoral votes for the Democrats, compared to 46% of the popular vote and 268 electoral votes for the Republicans. We also noted that a deeper analysis of these results showed an improvement in Kamala Harris's electoral prospects compared to Joe Biden's, with an average increase of 4.71% in the "swing states," highlighted by notable gains in states like Michigan and Wisconsin, where the increase reached 6%.
Finally, we pointed out that the "swing states" the Democrats were projected to win showed a solid lead in voting intentions: +2.2 points in Michigan, +2.2 in Pennsylvania, and +3.2 in Wisconsin. On the other hand, the Republicans, with the exception of North Carolina (+2.2), were defending Arizona, Georgia, and Nevada with very narrow margins of 0.2, 0.8, and 0.4 points, respectively.
This suggested that the likelihood of the Democrats increasing their number of electoral votes, and thus their number of "swing states," was greater than the likelihood of the Republicans winning in those states currently favoring the Democrats.
But what is the current situation now that we've reached September 1st?
Since August 17th (the date of the previous analysis) until today, the situation has been volatile, influenced by specific and significant events in the presidential race. In this second half of the month, two major events took place: the Democratic National Convention (DNC) and Robert F. Kennedy Jr.'s withdrawal from the presidential race.
These two events have caused fluctuations in voting trends, which we will now analyze:
Between August 1st and 17th, the Democrats maintained a 47% voting intention, while the Republicans were at 46%, as mentioned in the previous analysis. On August 19th, the day the Democratic convention was held (but before its conclusion), a variation was recorded: the Democrats' voting intention dropped by 0.5%, falling from 47% to 46.5%. However, the Republicans experienced a more significant decline, dropping from 46% to 44%, losing two points.
Subsequently, after the DNC, between August 19th and 22nd, the Democrats saw their voting intention increase by 0.8%, rising from 46.5% to 47.3%. The Republicans also showed a slight recovery, moving from 44% to 45.33% in the same three-day period.
The analysis of the DNC's impact was interrupted on August 22nd due to a significant event that occurred the next day: the withdrawal of independent candidate Robert F. Kennedy Jr. from the presidential race. By August 23rd, voting intentions stood at 48% for the Democrats and 46.5% for the Republicans, who had managed to regain some ground after dropping to 44% a week earlier.
From then until August 31st, the following variations were observed: The Democratic Party increased its voting intention from 48% on August 23rd to 48.27% on August 31st. Meanwhile, the Republicans saw a slight increase, moving from 46.5% on August 23rd to 46.6% on August 31st.
Looking at the trend line, it is evident that despite the Democrats' post-convention bounce, they have not managed to significantly pull away from the Republicans. The increase for the Democrats from the day of the DNC to the end of August was only 1.77% (from 46.5% to 48.27%), a negative indicator, as the average post-convention bounce typically ranges from 3 to 5 percentage points, well above the increase recorded by the Harris-Walz ticket.
This could suggest that the Democrats are approaching their electoral ceiling, which is not a positive sign. After the post-convention bounce, it is common for the increase in voting intention to level off, which could reduce support for the party in question.
One possible explanation for Kamala's lack of a significant surge after the DNC—resulting in a less pronounced post-convention effect compared to previous elections—could be the euphoria generated following her nomination as the primary candidate. This nomination might have led to a rapid and intense increase in voting intention before the convention even took place. This pre-convention surge could have anticipated the "explosion" that would typically be expected after the DNC but occurred much earlier.
Even if this explanation holds true, it remains a concerning sign for Kamala. The most logical interpretation for the absence of a significant post-convention boost, in addition to the increase following her nomination, is that she may be reaching her electoral ceiling.
Why is this a concerning sign? Why would a 48.3% popular vote be considered a poor ceiling?
The primary issue with this figure as an electoral ceiling lies not in the percentage itself, but in the margin relative to the Republicans.
Currently, Republicans have 46.6% of the vote, representing the second-best percentage achieved by Trump to date. Consequently, the current margin between the Democrats and Republicans is 1.61%.
Prominent political analysts, such as Nate Silver, have developed models that illustrate the necessary margins for each party to approach victory. According to these models, a gap of 1 to 2 percentage points gives the Republican Party a 68.7% chance of winning the presidency. Therefore, the current margin may not be sufficient to ensure a victory for Kamala Harris and leaves the Republicans very close to securing the presidency.
The elections of 2016 and 2020 clearly illustrate the importance of the vote margin relative to the opponent.
In 2016, Hillary Clinton received 48.17% of the popular vote, a figure very close to the current projection for Kamala Harris. Donald Trump, on the other hand, received 46.15%, which is also close to current projections but 0.45% lower, or nearly half a point. The difference between the two candidates was 2.02%. Ultimately, Trump won all the swing states and secured 304 electoral votes, while Clinton obtained 227.
In 2020, Joe Biden achieved 51.38% of the popular vote and 306 electoral votes, whereas Donald Trump, despite improving his previous results, received 46.91% and 232 electoral votes. In this case, the difference in the popular vote was 4.47%.
According to Nate Silver’s predictive model, with a margin of this magnitude, the Democrats would have approximately an 87% chance of winning the presidency. This is a significant contrast to the current situation.
But what is the situation in the swing states? Are the projections unfavorable for the Democrats? Let’s see:
The results that the polls show us in the swing states are as follows:
Arizona: Trump 48% - Harris 46.8% = Rep +1.2
Nevada: Trump 47% - Harris 48.75% = Dem +1.75
Georgia: Trump 46.75% - Harris 47.5% = Dem +0.75
North Carolina: Trump 47.75% - Harris 47.5% = Rep +0.25
Pennsylvania: Trump 46.67% - Harris 48.33% = Dem +1.66
Michigan: Trump 46% - Harris 48% = Dem +2
Wisconsin: Trump 45.67% - Harris 49.33% = Dem +3.66
Therefore, despite the doubts that the polls might raise regarding the popular vote, the Democrats still have very solid numbers in the swing state polls.
If we compare these polls to those from two weeks ago, we can observe the following:
Arizona: Trump 48% - Harris 46.8% = Reps +3 / Dems +1.8
Nevada: Trump 47% - Harris 48.75% = Reps +2 / Dems +3.75
Georgia: Trump 46.75% - Harris 47.5% = Reps -0.25 / Dems +0.5
North Carolina: Trump 47.75% - Harris 47.5% = Reps +0.75 / Dems +2.5
Pennsylvania: Trump 46.67% - Harris 48.33% = Reps +0.67 / Dems +0.33
Michigan: Trump 46% - Harris 48% = Reps +1 / Dems +1
Wisconsin: Trump 45.67% - Harris 49.33% = Reps -0.33 / Dems +0.33
These data indicate that, in the states that the Democrats were defending in mid-August, they still maintain an advantage over the Republicans, although with some variations. In Wisconsin, the lead has increased from 3.2 to 3.66 points. However, in Michigan, the lead has slightly decreased from 2.2 to 2 points, and in Pennsylvania, the difference has decreased from 2.2 to 1.66 points.
The main problem arises for the Republicans. They would go from defending Nevada, where they had a 0.4-point advantage, and Georgia with a 0.8-point advantage, to losing them, ending up 1.75 points behind in Nevada and 0.75 points behind in Georgia, which would now be defended by the Democrats.
In North Carolina, their lead has decreased from 2.2 points to only 0.25 points, which means they would have lost 1.95 points of lead. The only state where the Republicans have increased their lead is Arizona, where they have gone from a difference of 0.2 to 1.2 points.
The map that these results would leave is as follows:
As we can see, Kamala’s victory at the end of the month is much more comfortable than it was 2 weeks ago. However, this is not as overwhelming as it might seem.
In Michigan and Wisconsin, the Democrats regained the lead since Biden expressed his support for Kamala Harris’s candidacy. This suggests that the low electoral expectations observed in these states in mid-July were due to the low enthusiasm towards the previous candidate. However, once the enthusiasm was recovered, the voters who initially considered not supporting the Democratic party due to their discontent with Biden would have returned. Therefore, barring surprises, there should not be much competition in these states on election day.
In Nevada, where the Democrats have achieved victories in the 2016 and 2020 elections, they currently defend a lead close to two percentage points, similar to the one they maintain in Michigan. Their electoral support stands at 48.75% of the total votes, a percentage that surpasses the one obtained in 2016 (47.9%) and places them just 1.35 points away from reaching Biden’s result in 2020 (50.1%). It is relevant to highlight that, since 1976, the party that exceeds 48% of the votes has won this state, and the Republican Party has not won a presidential election in Nevada since 2004.
What I want to say with this is that these states are not decisive in this election (barring surprises, I reiterate), nor do North Carolina (which has voted for the Republican Party since 2012 and, in 2020, the difference was more than 1 point) or Arizona, which, despite being one of the states that did change color between the 2016 and 2020 elections, seems not essential to secure the presidency.
In my view, the two decisive states, that is, where the election is really at stake, are Georgia and Pennsylvania.
Let’s look at the following map:
This map projects a Democratic victory in the states of Nevada, Wisconsin, and Michigan, which would give them a total of 257 electoral votes. On the other hand, a Republican victory is projected in Arizona and North Carolina. This analysis is based on the previously mentioned criteria and reveals the following: Arizona is not a crucial state. If the Democrats won in Arizona, they would be only two votes away from reaching the 270 needed, but they would still depend on a victory in Georgia or, ultimately, in Pennsylvania to secure the presidency.
In particular, if the Democrats lost Arizona but won Georgia, they would reach 273 electoral votes, enough for Kamala Harris to become President. If they won in Arizona but lost in Georgia, they would need to win in Pennsylvania to secure the victory. If they lost both Arizona and Georgia, they would still depend on a victory in Pennsylvania. Finally, if the Democrats lost in Arizona and Pennsylvania, they would still need to win in Georgia.
For the Republicans, Arizona is also not essential. However, it is crucial for them to win both Georgia and Pennsylvania to secure the presidency.
Therefore, the fundamental states in this election are Georgia and Pennsylvania. While the Republicans cannot win without achieving victory in both states, the Democrats need to win at least one of the two to secure the presidency.
A final observation concerning these states is that, over the past month, the advantage previously held by the Democrats in Pennsylvania, as noted earlier, has diminished by more than 0.5 points. Furthermore, although the situation in Georgia has reversed with the Democrats now leading, their margin over the Republicans closely mirrors the margin the Republicans previously held when they were in the lead. This suggests that the volatility of this state leading up to the election day is considerable and indicates that the margins are likely to remain narrow. Therefore, the outcome in Georgia will largely be determined by the independent voters.
To conclude this analysis, I would like to add, briefly, that in the upcoming elections there could be very clear indicators about which side the victory will lean towards.
If the Democrats lose Nevada, their chances of losing the presidential race would increase exponentially. Similarly, for the Republicans, a loss in North Carolina would be an almost unequivocal sign that they will lose the presidential race.
Next week, we will address the importance of unwritten norms for maintaining solid democratic health. In 15 days, we will meet again to analyze the results and trends offered by the polls in mid-September.