Why The Polls Keep Getting It Wrong

Friends —

I worked at Fox News during the 2016 election, and I can tell you firsthand: most people there were as shocked as anyone else when Trump won. Every polling model called for a comfortable Hillary Clinton victory, and to the experts, Election Day was almost a formality.

That memory has stuck with me for a decade. Because it wasn’t just one bad night. It happened again in 2020. The polls actually got worse underneath, even though Biden won. Then came 2024, and the industry missed again, this time in every single swing state.

So I wanted to dig into what actually happened inside the polling industry. Not too much of the “rigged election” stuff (but a little), and not the “shy Trump voter” explanation either, which I’ll show you doesn’t hold up. The real story involves the move to cell phones, collapsing response rates, and an industry that knew its foundation was cracking years before 2016 and didn’t solve it.

On top of those problems, their attempts to fix things allowed their biases to creep in.

When everyone you work with and socialize with shares the same assumptions about what’s politically possible, those assumptions start feeling like facts. I think that explains a lot about why an industry full of smart people with unlimited data kept getting this wrong, year after year.

This video digs into the mechanics most people never think about — and I think once you see it, you’ll never look at a poll the same way again.

– Ken

PS: I am soon heading to El Salvador. It went from one of the worlds most crime-ridden countries to one vying with Canada as the safest in either North or South America. I find that intriguing.