There have been a lot of recent statements essentially making claims about base consolidation on the Republican side of the aisle. Almost all of the claims I've reviewed are inaccurate, incorrect, demonstrably false.

We've been tracking Trump or Bust voters alongside normal gauges pertaining to base consolidation for each candidate, for months. At the heart of these false claims is a lack of understanding RE: proportionality.

In order to better illustrate proportionality, I created the chart for "Republican Candidate-Specific Support (December 2023)" based on the survey question bearing the same name asked of 1,339 eligible Republican primary voters (and caucus-goers) interviewed for the National Poll in December.

It very much demonstrates and supports what Harry Enten attempted to argue in a recent column, which is that Donald Trump is by far the most likely to 1) consolidate the Republican primary coalition in a general election and 2) drive out the largest number of voters.

Take, for instance, the claim from Haley supporters that her voters could very well back Joe Biden in a general election. Indeed, 20.7% of her voters indicated in our poll that they would consider doing just that—and other polls mirror our findings. However, while "1 in 5 Haley voters would consider backing Biden" sounds like an alarming headline for the eventual nominee, the actual number of voters (27) is less than the 3.8% of Trump's voters (32) who indicated they would consider doing the same.

The 21.2% of Haley voters who would consider voting for a third party candidate is far less real votes (28) than the seemingly smaller 7.8% share of Trump voters who said the same (66). That's the impact of proportionality.

Put simply, Trump share of the vote is far larger, thus it is far more difficult (less likely even) for non-Trump candidates to consolidate his support if he is the nominee than visa-versa, and certainly they are more risky.

Further consider, the 1.7% of DeSantis supporters who said they would not vote at all if their candidate was not the nominee was derived from just TWO interviews. Two, that's it. By comparison, the 3.7% of Trump supporters who said the same represents interviews from 31 people.

Of course, the real risk and potential serious damage to any non-Trump coalition comes from the 39.7% of Trump voters who say they'll write-in his name if he's not the nominee. That means of the 1,339 total primary voters we interviewed, 334 explicitly stated they'd opt for Trump as a write-in protest vote.

Most conclusive, the total number of voters who indicated primary support for another candidate other than Trump and also indicated they might not vote for the party's nominee (149) is less than half the number of voters who simply say they would write-in Trump (334), and dwarfed further by the total number of Trump supporters (463) who indicated they might not vote for the party's nominee.

Hopefully, the colored columns demonstrate the impact of proportionality better than I, or anyone, can explain it.
https://www.datawrapper.de/_/PWwza/