What if America used ranked-choice voting for presidential primaries?

Dear everyone,

Welcome to a special Thanksgiving week edition of the Strength In Numbers monthly Q&A! Thanks to everyone who submitted items for this month’s installment. I have selected eight questions that were most topical and that will provide the most value for the full audience. You can send questions for next month to me via email (questions[AT] gelliottmorris< DOT>com) or leave them in the comments of this post.

In this installment, I answer reader questions like...

• How do your polling average models work?

• What would ranked-choice voting in the Democratic Party look like?

• Are election results driven more by turnout or persuasion?

• How do you ensure your biases don’t influence your work?

• Do non-voters still lean toward Trump?

Let’s dive straight in. As usual, the Q&A will be paywalled about halfway through as an exclusive for paying members of Strength In Numbers. Premium subscribers get tons of benefits, including exclusive analysis weekly and early access to new data products. You can sign up to read the full version here:

Does a universe of “all potential voters” still lean Republican?

Doug asks:

I often see people comment that, had the tens of millions of folks who didn’t vote in the 2024 presidential election submitted a ballot, then Harris would have won. Are there data that support this? I seem to recall a poll (maybe by Pew) earlier this year which found the opposite to be true, that Trump would still have won.

Apologies if you have covered this ground before (I’m a pretty new subscriber) and thank you for all that you do to present and interpret polling data.

First, thanks for joining us here at SIN, Doug!

My answer is that had everyone voted in 2024, Trump likely still would have won. As Doug mentions, an analysis of voter file and survey data from the Pew Research Center in early 2025 found the opposite. According to Pew, survey respondents who did not vote in 2024 favored Donald Trump by four points, 44% to Kamala Harris’s 40%.

This was a big shift from previous elections, when non-voters favored Democrats. Hillary Clinton’s lead was large enough with non-voters, for example, that she very likely would have won in 2016 if everyone had turned out.

But it looks like opinion has shifted since 2024. In August, Trump’s job approval rating among 2024 non-voters was down to 32% compared to 45% ...