As California’s 30-day (ish) recall election proceeds to “Election Day” on September 14, there’s a new poll almost every day. They’re generally consistent, with a few outliers – as usually happens. A poll released Wednesday by The Trafalgar Group, though, is raising some eyebrows.

#CARecall – @GavinNewsom remaining Gov. has significant lead.

No, keep him 52%,
Yes, remove him 44.4%,
Unsure 3.7%,

While @larryelder leads by a plurality if recall succeeds – according to @trafalgar_group #CA #Recall #Poll conducted 8/26-29 See Report: https://t.co/oA0LIFe4aw pic.twitter.com/IELqWeqbPj

— Robert C. Cahaly (@RobertCahaly) September 1, 2021

Trafalgar, touted as a “Republican” firm and generally trusted by those on our side as reliable, showed Newsom beating the recall by 8 points and Republican Larry Elder as the leading replacement candidate, neither of which are surprising. What was surprising was the inclusion of Doug Ose (who withdrew from the race due to health concerns 10 days before the survey was conducted and endorsed Kiley) and the exclusion of Asm. Kevin Kiley (who has polled either 3rd or 4th in all of the other polls and is one of the three Republicans who participated in every televised debate).

Also, Democrat/YouTuber Kevin Paffrath performed similarly in the first Survey USA poll, conducted a few weeks ago, and in Survey USA’s follow-up, released the same morning, Paffrath’s support was down at 6 percent. Here’s why, according to the firm:

SurveyUSA today names all 46 candidates; in previous polling, SurveyUSA named 7 viable candidates only, including a single Democrat, Paffrath. When respondents today are shown the full list of candidates, which includes 9 Democrats, Paffrath’s support reverts to what may be its truer level, 6%.

I was curious whether Trafalgar gave the respondents a full list of candidates or asked an open-ended question, or if there was a question where party affiliation wasn’t listed – because if only one Democrat was on the list of responses, it stands to reason that Democrat would poll on the high end. And, wondering about the survey design and who sponsored it, I clicked through to Trafalgar’s site.

According to the seven-page document (consisting almost entirely of graphs) labeled as the “full report” on Trafalgar’s website, the poll of 1,088 “likely ballot casters” in the 2021 recall was conducted between August 26-28, 2021, and claims 95% confidence and a 2.98% margin of error. The methodology is not listed; instead, viewers are referred to a page on Trafalgar’s website: TheTrafalgarGroup.org/Polling-Methodology. The methodology on that page isn’t specific to this poll and, really, isn’t very specific at all.

Judging by the report, the only candidates presented to survey respondents when asked the question, “Should Newsom be recalled, who, of the following candidates, would you most likely support to replace him?” were Elder, Paffrath, Kevin Faulconer, John Cox, Caitlyn Jenner, and Doug Ose. Respondents could also say they were undecided or “someone else.”

I also asked Trafalgar (via tweet) who paid for the poll.

No one paid for it. We did it as we do all horse race polls for publicity. We get lots of business because we usually have the lowest error from vote result than other firms.

— The Trafalgar Group (@trafalgar_group) September 1, 2021

They replied:

“No one paid for it. We did it as we do all horse race polls for publicity. We get lots of business because we usually have the lowest error from vote result than other firms.”

It still didn’t make sense, especially because I had already been contacted by a source claiming to have been involved with the poll, who said that Kiley was specifically excluded from the poll and that the vast majority of the people who replied “someone else” mentioned Kiley by name. Since “someone else” came in at 9.1 percent, that means that Kevin Kiley was the Republican in second place. And, if Trafalgar was doing it for publicity and ostensibly would use the results in marketing, wouldn’t they want to be on record with a poll showing Kiley with 9-ish percent support in case that’s how it actually turned out?

Not to mention the fact that everyone who’s been closely following the recall knows that Elder’s main competition is Kevin Kiley. Longtime activists, frustrated by Elder’s refusal to even appear in the same room as Kiley and by his refusal to appear before their grassroots clubs because they can’t promise big contributions, are struggling to decide which man will get their vote. To leave Kiley out of a horse race poll and promote the results as objective is professional malfeasance.

While picking my jaw up off the floor, I received a Twitter direct message from Trafalgar. I was informed that questioning who paid for [the poll] wasn’t fair, that they always disclosed if they were paid to conduct a horse race poll, and that they would “address the Kiley issue” by conducting another poll with Kiley included.

I then received a Twitter direct message from Robert Cahaly. He stated that they had originally run the poll for a private client who didn’t want it released and Kiley didn’t show up in the top 10, then they ran one themselves for media and didn’t update the top 10. He added that it’s likely that the few GOP who stated “someone else” were for Kiley. Unfortunately, we don’t know how many GOP who stated “someone else” there were, because there isn’t enough information provided about the poll to make that determination.

There are some major issues with this sequence of events and some assertions that simply don’t stand up to scrutiny.

First – and I’ve worked on numerous campaigns, so I know this firsthand – if a campaign hires a pollster for an internal poll, and they don’t want those results released to the public, if that pollster turns around and duplicates the poll on their own dime and releases the results, the campaign would be livid and likely never use that pollster again. There’s a reason they didn’t want those results released to the public. If they did, and they didn’t want it coming from them, they’d find a way to leak the results. That’s just how it goes.

Second, why would Cahaly admit to sloppy poll design? Does someone who’s considered to be one of the best pollsters in the country simply forget to “update the top 10”? That wouldn’t give me a hell of a lot of confidence in the work product.

Third, no matter how Cahaly defines “top 10,” Kiley would have been in it. In every other poll he’s in the top 3 or 4 going on percent support; oftentimes two or three candidates tie for second or third.

Fourth, where is the mention of “top 10” in the survey design? It’s completely absent. Were there precursor questions that we don’t know about in the design?

Fifth, it’s completely unethical to lie/mislead about who sponsored a poll. I understand Cahaly is claiming that the poll results he released weren’t sponsored by anyone, but that’s splitting hairs. He re-ran the exact poll he ran for a private client and released those results. That means that this private client sponsored the poll – and I don’t believe that Cahaly ran the exact same poll a second time. It just doesn’t make sense. The private client instructed Cahaly to exclude one of the leading candidates for a reason. What was that reason? Was it to suppress the Kiley vote? Was it to make it look like there really is no competition on the GOP side for Elder, and that it’s really an Elder/Paffrath race? If we knew who paid for it, we’d know the answer to that question.

I don’t believe that Larry Elder’s campaign had anything to do with it, because this poll isn’t a total positive for him. It energizes Newsom’s base, who are more afraid of a Trump-supporting Republican becoming California governor than literally anything.

Let’s hope that Cahaly is out with a new, properly-designed poll including Kevin Kiley soon since every day more and more people are casting ballots and some are undoubtedly influenced by a poll from a firm as reputable as Trafalgar (as the social media responses attest). At the very least, Cahaly owes California voters the truth about who tried to sway the recall election by disseminating intentionally misleading poll results.

(NOTE: RedState’s California authors have unanimously endorsed Kevin Kiley for governor.)