Dixon’s push poll, continued

News and analysis by Geoff Kelly, Investigative Post's political reporter

A couple Sundays ago, Buffalo News politics columnist Bob McCarthy wrote, in essence, that I got it all wrong in my June 19 piece about Lynne Dixon’s push poll.

I don’t know that I did. I don’t think McCarthy knows that I did, either.

Dixon is challenging incumbent Erie County Executive Mark Poloncarz. In early June, her campaign commissioned a poll of 1,325 likely voters.

I first read about the poll in McCarthy’s June 16 story, which touted the Dixon campaign’s conclusion that the race was a statistical tie — a good sign for the underdog. Naturally, I wanted to know more about the poll. I knew eventually the Dixon campaign would file the poll with the New York State Board of Elections, as required by state law, but I didn’t want to wait.

So I tracked down a few people who had taken the automated phone survey. It looked to me, based on the conversations I had with poll participants, that respondents who expressed support for Poloncarz were led into a menu of questions intended to cast Poloncarz in a negative light. A classic push poll.

Those push poll questions did not appear in the poll that the Dixon campaign filed on June 18 with the state Board of Elections. Nor were they in the poll the campaign provided me, when I asked for it, the same document on which McCarthy had based his story. 

I took issue with McCarthy’s story, and his editors, for not looking deeper. After all, there was chatter on social media about the push poll questions even before my column describing them appeared. Highlighting the happy conclusions of the Dixon campaign, without investigating the underside of the poll, gave the campaign favorable coverage that it could (and did) re-package for social media and mailers.

A couple weeks later, McCarthy called me. He’d been right, he told me. And I’d gotten it wrong.

He told me he’d discovered the Dixon campaign had commissioned two separate polls. One was the “straight” poll, the one on which he’d based his story — the one eventually filed with the state Board of Elections. The second was the push poll, which he said he had not known about because the Dixon campaign had not revealed it to him. Because the Dixon campaign had not discussed that second poll, or released its results, the campaign had not been obligated to file it with the state Board of Elections.

I wondered how he knew there was a second poll. 

I’m still wondering. 


iPost benefit: Tom Toles & The Outlyers, Aug. 16 at Sportsmen’s


When I read McCarthy’s July 7 column, I’d expected evidence: an invoice or a contract with the polling company, for example. Perhaps even a copy of the poll’s script, though providing that would have required the campaign to file the alleged second poll with the state Board of Elections. That’s something the Dixon campaign clearly would rather not do, and still has not done.

Instead, McCarthy seemed to be taking the word of Chris Grant, CEO of Big Dog Strategies, the principal consultant for the Dixon campaign. At least, that’s the only evidence he presented in the column: a conversation with the former chief of staff to Congressman Chris Collins.

Chris Grant,  whose house was raided by authorities in an investigation that also swept up former Deputy Mayor Steve Casey and political operative Steve Pigeon. Only Pigeon was eventually charged.

The same Chris Grant who was questioned (but again not charged) by investigators looking into the insider trading case brought against Collins. 

Chris Grant, the author of too many truth-bending and wickedly effective political mailers and ads to count.

Chris Grant is  a hard-charging political operative, no doubt, with a reputation for winning. But if he tells me the sky is blue, I reach for an umbrella — just in case.

In fact, Chris Grant hasn’t told me anything. In the past three weeks, I’ve asked Big Dog Strategies over and over to provide me some evidence, any evidence — a contract or an invoice — that the campaign commissioned two distinct polls. No response.

I sent messages to Co/Efficient, the Kansas City, Missouri firm that conducted the poll. No response.

I asked McCarthy again how the Dixon campaign convinced him there was a second poll. His reply:

“If the Dixon campaign filed the ‘straight poll’ with the BOE but asked the ‘push poll’ questions, then they violated state election law. If that’s the case, you’ve got a great story.”

True, but it doesn’t address my question. 

Our Weekly Newsletter

If there was just one poll, as McCarthy points out, then what Big Dog Strategies filed with the state Board of Elections was a lie. But the state Board of Elections have told both me and McCarthy that the Dixon campaign is not obligated to file an alleged second poll — the push poll — not even to prove that it exists separate from the poll already filed.

So we find ourselves stuck in a Catch-22: All my conversations with people who actually took the poll — that’s up to seven now — suggest there was just one. But of course that doesn’t prove there was just one. 

Nor is there any actual proof that there were two, at least none that’s been presented publicly. It’s just Grant telling McCarthy a story that exonerates the Dixon campaign of potential wrongdoing. 

The only way we’ll ever know the truth is if the Dixon campaign provides evidence. They are unlikely to do that unless compelled by the state Board of Elections. And that won’t happen unless someone brings a formal complaint against the Dixon campaign, alleging a violation of election law.

Grant’s word is not enough for me. Given my previous reporting and the evasion of the Dixon campaign, I remain confident my column was correct.