James Fishback, who earned just 10.5 percent in Florida’s Republican gubernatorial primary, released a survey Monday claiming 28 percent of GOP primary voters would back Democrat David Jolly in November — and that 70 percent of Republican primary voters ages 18–29 would do the same. He framed it as proof that nominee Byron Donalds is “failing to consolidate” the party and that Florida is “on track to elect its first Democrat governor in three decades.”

Ben Galbraith, senior pollster at Victory Insights, called the survey “complete BS” and recommended muting Fishback “unless you’re interested in losing a few IQ points every time he posts.” In a follow-up, Galbraith laid out the sampling problems that make the numbers unserious.

The poll of roughly 3,000 self-identified Republican primary voters included 1,252 who said they backed Fishback, 390 who backed Jay Collins, 61 who backed Paul Renner, and 42 who backed Bobby Williams. That is roughly three times as many Fishback voters as Collins voters — even though Collins received about 2.4 times as many actual primary votes (427,000 to Fishback’s 178,000). It also includes about 20 times as many Fishback voters as Renner voters, though Renner finished just two points behind Fishback (8.5 percent to 10.5 percent).

A sample that heavily overweights the candidate who just lost is not a picture of the Republican primary electorate. It is a picture of Fishback’s few remaining supporters. Those voters are the ones most likely to tell a pollster they are still angry at Donalds. Treating that group as representative of all Republican primary voters is how you manufacture a 70 percent youth defection number.

There are other issues with Fishback's poll. No pollster is listed, suggesting that Fishback likely conducted the poll himself, raising serious questions about conflicts of interest and methodological rigor. It's unclear how Fishback fielded the survey and where he sourced the contacts from.

This is not Fishback’s first polling controversy. During the primary, The Public Sentiment Institute released surveys that showed him far more competitive than he proved to be. Various polling aggregation sites later barred the firm after finding it had failed to disclose Fishback campaign payments and had “leaned” some Collins respondents toward Fishback because the raw data “did not align with [their] read of the race.” Fishback finished third, well behind Donalds (47.8 percent) and Collins (25.2 percent).

Brendon Leslie, Founder & CEO of the influential Florida's Voice news outlet, called Fishback out for his pattern of flaunting polling numbers from dubious sources. He accused Fishback of working to get socialists elected.

The new survey continues the same pattern that Fishback has leaned on: produce a headline that hurts the nominee, then dare Republicans to ignore it. Florida’s general-election polling is competitive, not collapsing. Recent surveys have shown Donalds ahead of Jolly, consistent with a red-state race that will still require turnout and message discipline. That is not the same thing as 28 percent of Republican primary voters already defecting.

Fishback continues to divide the GOP base. Earlier this month, he held an extremely limited gasoline sale at "pre-Iran War prices" in protest of President Trump's military actions in Iran. Nevertheless, after his resounding primary election loss, he's beginning to slide further and further into irrelevance as the Republican base consolidates behind Donalds ahead of the general election in November.