(War Room) On Wednesday, political analyst and pollster Mark Mitchell and WarRoom host Peter Navarro explored the intricacies of Rasmussen polling data and its implications for the upcoming 2024 presidential election. Their conversation highlighted significant contrasts between polling organizations and the evolving political dynamics over the past four years.
Mitchell took aim at corporate- mainstream polling firms, specifically criticizing 538 for alleged methodological biases. “538 is owned by ABC News, which rolls up to Disney,” Mitchell stated, arguing that the aggregator’s approach has shifted to align with particular narratives. He contended that, unlike 538, Real Clear Politics offers a more transparent view of the polling landscape.
Mitchell criticized the treatment of his polling data by major aggregators like 538, which he claims have acted unfairly toward his work. According to Mitchell, 538, has shown bias in their reporting and even threatened to exclude his polls from their aggregation. He said, “I think they’ve kicked us out quite frankly. I think that’s the reason.” Mitchell expressed frustration that his polling data, which he believes offers a more accurate picture of the electorate, is not being fairly represented by these aggregators.
Mitchell contrasted his methodology with that of 538, arguing that 538’s models are opaque and designed to drive narratives, while his own approach aims for maximum transparency. He emphasized that his polling is grounded in straightforward, consistent methodologies without selective adjustments: “I didn’t change anything… I’m basically trying to match what I think an accurate picture of the electorate is.”
This critique exposes a broader debate about polling accuracy and transparency. Mitchell’s data suggests that Kamala Harris’s approval ratings have been largely stagnant despite media portrayals of rising support. “Kamala Harris’s numbers have been stagnating around 44-45%, contrary to media reports of improvement,” Mitchell noted.
His skepticism extends to manipulating polling methodologies, alleging that some organizations adjust their methods to achieve desired outcomes. He specifically pointed to recent admissions by polling firms about over-sampling certain demographics, which he believes distorts the true picture of public opinion.
The discussion also ventured into the shifting political landscape compared to the 2020 election. Mitchell highlighted that while Trump’s numbers are competitive, they vary significantly across different battleground states. “Trump is roughly tied within the margin of error in states like Arizona, Michigan, and Pennsylvania,” he explained.
This variation indicates a more volatile electoral environment than in previous cycles. Despite some recent polling suggesting tighter races, Mitchell observed a notable improvement in Trump’s standing compared to four years ago. “Trump’s performance is better compared to 2020, showing a significant pickup in key states,” he emphasized.
This improvement, however, comes with caveats. Mitchell noted that while Trump’s current position is stronger than before, discrepancies remain in certain states. For instance, in Arizona, where Trump’s numbers are not as favorable as they were four years ago, he suggested that methodological issues might be at play.
“I think I got a pretty left sample in Arizona that didn’t fully get weighted out,” he speculated, reflecting concerns over the accuracy of state-level polling
