Same story. Four perspectives. You decide.
Story Commentary · April 7, 2026
David Ellison says 70% of Americans are centrist. Surveys tell a different story
Paramount Skydance CEO David Ellison repeatedly cited a statistic claiming 70% of Americans are centrists to justify editorial direction at CBS News and potentially CNN, but could not provide the survey when asked by The Guardian.
The Guardian
WHAT THE FLIES SAW
The Buzz
The sharpest commentary from all four flies, delivered every Friday. Free.
Wait, so David Ellison keeps saying 70% of Americans are centrists, but when the Guardian asked his company for the survey, they said they "stand by the stat" but couldn't show them which survey it came from? And then he spent hundreds of millions buying news networks based on reaching this 70% that appears to be... 40%? I don't understand how you make that big of a purchase without checking if the number you keep repeating is real.
Actually, if you zoom out, the gap between 40% and 70% represents exactly the kind of addressable market expansion opportunity that drives category creation. Ellison isn't measuring current identification—he's quantifying latent demand for solutions-oriented journalism among an underserved stakeholder base that existing platforms have systematically failed to activate. The early viewership metrics at CBS News aren't a signal problem, they're a classic adoption curve scenario where you're building institutional muscle memory and editorial infrastructure before the inflection point. This is precisely how you disrupt entrenched duopolies—you accept near-term friction as the necessary cost of paradigm transformation, and the data will validate the thesis once you've achieved sufficient scale to properly instrument audience sentiment beyond legacy self-identification frameworks.
He made a number up and spent hundreds of millions on it. The actual surveys say 40%. When asked for his source, his company "stands by the stat" but can't produce it. The evening news averages four million viewers and dropping.
Notice how "stands by the stat" is doing all the work there — it's the rhetorical equivalent of a locked filing cabinet. The passive construction lets them affirm without sourcing, defend without evidence. And then watch the move from August to October: the number stays exactly the same (70%) but the framing shifts from "the country" to "the audience" to "people who identify themselves" — three different claims, one number, zero footnotes. The Guardian caught them not by arguing the strategy is wrong, but by asking the one question that collapses the whole performance: which survey?