Blog/Analysis
The Mood of the Swing States: U.S. Election Analysis from Focaldata & Signal AI
Today’s Landscape: Media Fragmentation and the Fight for Voter Attention
If you were forced to describe this year’s U.S. Primary election in one word, “whiplash” might suffice. The back-and-forth media mayhem of this year’s race has been dizzying—from the incumbent president, Joe Biden, unexpectedly dropping out in July to the swift nomination of Kamala Harris to the attempted assassination of Donald Trump—all within the same month. It feels impossible to keep up with the torrent of news, opinions, polls, and predictions. Amid this frenzy, how can we assess the true mood of voters, particularly in key swing states? And how can we determine which candidates are actually making progress in this volatile landscape?
As we sift through the noise of campaign coverage, it’s becoming clear that we are no longer getting our election news from a single, reliable source. Axios recently coined the term “shards of glass” to describe our fragmented media ecosystem—there’s no longer one source of truth; instead, we have countless outlets offering different versions of reality. In this chaos, voters are left trying to make sense of conflicting reports and shifting narratives.
To help cut through this clutter, Signal AI, a reputation and risk intelligence company, partnered with research powerhouse Focaldata to dive deeper into the election data. By combining cutting-edge artificial intelligence with robust voter polling, we aim to paint a clearer picture of how voters really feel—and which campaign strategies are resonating.
Signal AI’s U.S. Elections Index uses big data and AI to analyze billions of unstructured media data points, distilling them into insights on coverage across critical metrics like share of voice, sentiment, and polarity. Meanwhile, Focaldata harnesses proprietary tech to collect data directly from verified respondents at speed. Their deep polling work spans the eight most critical swing states, providing a real-time pulse on voter sentiment.
The results of this collaboration are revealing—and, in some cases, surprising.
Candidate Favorability: What Voters Think vs What the Media Says
We first used our respective data sources to examine favorability toward the two main candidates. Signal AI’s sentiment analysis within the US Elections Index generates a score based on the split between positive and negative media coverage. Focaldata, on the other hand, asked a representative sample of voters in each swing state what they thought of Trump and Harris and calculated a net favorability score based on the results.
The chart below plots these scores in each state for both candidates, with the darker dots representing the Focaldata scores and the lighter ones representing the scores produced by Signal AI.
When comparing media sentiment to polling data, one of the most glaring discrepanciesis the difference in candidate favorability. According to Signal AI’s sentiment analysis in the Elections Index, Harris consistently outperforms Trump across media outlets in every swing state, though only by narrow margins. We can also see that the overall sentiment of media coverage hovers near a neutral sentiment score across the board.
However, Focaldata’s polling tells a different story: Trump leads Harris in three of the seven swing states, including a significant (and perhaps unbridgeable) eight-point advantage in Ohio and North Carolina. Political science research shows that leadership ratings are highly predictive of election outcomes, so it’s no surprise that, in Focaldata’s vote intention polling, Trump’s chances of taking Ohio are also ranked highly (and outside the all-important margin of error).
The disconnect between media coverage and voter perception is stark—and it highlights the complexity of predicting outcomes in such a fragmented information environment.
Polarization Scores: The Dividing Lines
Are there any undecided voters left? It’s no surprise that the race is polarized.
Polarity is a measure of how split attitudes are across the two candidates. In other words, it scores the balance of positive and negative conversation: 100% polarity means a total stalemate—there are 50% positive and 50% negative attitudes towards a candidate. By contrast, 0% means opinion exclusively favors either one of the candidates. Again, in the chart below, the Focaldata points are shown in the darker circles whereas Signal’s scores for both candidates are in the lighter shade.
Overall, opinion toward both candidates sharply split across the polling samples, with polarization scores high for Trump and Harris. However, the media analysis results were more varied.
Trump’s media coverage is remarkably one-dimensional in Ohio, where he has a polarization ‘floor’ of just 16% of coverage showing any significant divergence in opinion. This underlines how one-note the media commentary is around him in the state.
On the other hand, Harris is seen as a far more polarizing figure across all states, suggesting that her media presence generates more animated reactions–both excitement from the left and animus from the right–even in states where she holds an advantage. The data indicates that even though Harris is the sitting Vice President, her last-minute nomination has accelerated scrutiny from both sides of the spectrum.
Issue Areas: Strengths and Weaknesses
Finally we look at the hot-topic issues shaping the election – again through the lens of polarity. The issues clustered in the top right-hand corner of the graph are the issues that scored highest across both data sources. For Focaldata, this represents the issues on which voters are most most undecided on which of the two main candidates ‘own’ the issue. For Signal AI, the score represents the issues on which there is the largest disparity in positive and negative mentions.
Each candidate’s strengths and weaknesses on policy issues are apparent. In both Signal AI’s media data and Focaldata’s polling, the top issues are jobs and the economy.
Perhaps that’s why both candidates are doubling down on dollars in the final weeks. Harris recently spent time in New York City meeting business leaders and announcing her commitment to a “bold, capitalist vision for the economy.” Trump is promising to trim taxes everywhere–whether for retirees’ social security benefits or baristas’ tips.
The least polarizing issues are national security and race relations. While those two are long-simmering political fault lines, recent bipartisan consensus on China and women’s rights post-Roe v. Wade’s overturn appears to have somewhat moderated the debate.
Interestingly, Signal AI shows a scattering of other issues high on the polarization index, like inequality and abortion, continuing to energize Harris’ base. However, these issues are so ‘locked in’ by Democrats from voters’ perspectives that we find a low polarization score with these on the polling side.
So, What’s the Strategy?
Traditional strategies are evolving in a campaign season marked by rapid shifts and a fragmented media environment. Candidates can no longer rely on singular messaging or conventional platforms; rather, they’re moving towards “unorthodox” approaches like podcasts. Instead, Harris and Trump must engage voters across multiple touchpoints—TV, podcasts, social media, and in-person rallies—while battling the barrage of conflicting media narratives. The challenge for candidates is clear: cut through the noise and connect meaningfully with voters in real-time.
Today’s political landscape mirrors the challenges corporations face in a similarly fragmented marketplace. Like political campaigns, businesses must navigate a constantly shifting media environment where consumer attention is scattered across countless platforms. The days of relying on a single, unified brand message are over. Instead, companies, like political candidates, need to adopt a multi-channel approach to reach their target audience. They must be agile, respond to feedback quickly, and adapt their messaging to resonate with different groups while maintaining a consistent core identity.
Understanding public perception through data-driven insights is now the key to success for both campaigns and corporations. Just as political candidates use AI to track sentiment and adjust their strategies in real time, companies must leverage analytics to understand consumer behavior and make informed decisions. In an era where the media landscape is fragmented and fast-moving, the ability to react quickly and strategically is critical. The brands and candidates that can do this most effectively will be the ones that not only survive but thrive.
In politics and business, the message is clear: mastering the art of real-time engagement, powered by data and AI, is the clear path to winning.
Stay connected
Subscribe to get the Focaldata AI newsletter delivered directly to your inbox.