One of the biggest obstacles for any independent or third party candidate is being ignored by the polls. The media, debating commissions, and voters all look at polls to assess any race, and candidates who don’t appear in them struggle to seem relevant.

When a candidate does not appear in the polls, they can expect to be excluded from almost everything.

Two unusually strong third party candidates are currently running for governor of New York: Larry Sharpe (Libertarian) and Howie Hawkins (Green). They have both been running well-organized campaigns for months, but polling companies are not including them in their surveys. Predictably, the media is mostly ignoring them.

When the Sharpe campaign grew frustrated with this and funded its own private poll at a cost of $5,000, they found some remarkable results: both Sharpe and Hawkins were sitting at roughly 5% each. That’s not winning, but its significant, and enough to have a real impact on the results.

It was a clear-cut example of polling organizations leaving out credible third party candidates even when their presence in the race was clearly important. I wanted to know why.

My colleague at News Growl, Patricia McConnell, and I contacted three polling organizations that cover New York State politics: Siena, Marist, and Quinnipiac, and asked them why Hawkins and Sharpe are being excluded. I was not confident we would get any replies, but I was pretty sure if we did we would just get a series of technical justifications. Perhaps including more than two options complicates statistical analysis, or maybe having candidates with vote shares smaller than the margin of error is problematic. I was expecting replies similar in tone to what I get from my accountant.

We did hear from two pollsters, Siena and Marist, and I was dead wrong about what they would say. To be honest, I was a bit shocked.

On the record, Siena admitted that they only cover Republicans or Democrats until late in the race. Their reasoning was the status of third party candidates are in “flux” until later in the race. That is exactly the opposite of reality. Third parties typically decide their candidate months before major parties, but they declined to alter their original comment.

It was a blatant case of institutional bias, but the honesty and straightforwardness was at least refreshing.

But it was Marist’s answer, and the implications that followed from it, that really stunned me, however.

Marist Poll works with media partners like NBC and the Wall Street Journal, and those partners ultimately get to choose who gets included in the polls and who does not. When it comes to primary races, especially in the early stages, they choose candidates based on a “news judgment.”

“News judgment” is another way of saying “newsworthy,” which is another way of saying how many people will want to watch or read a story. Newsworthiness is a legitimate journalistic concern, but it also sells newspapers and bumps up ratings.

A case in point is a Marist Poll from the last New York Governor’s race, published in March 2014. It includes head-to-head matchups between Mario Cuomo and three potential Republican opponents: Rob Astorino, Carl Paladino, and… wait for it… Donald Trump!

Yes, Donald Trump.

Did you know Donald Trump flirted with running for governor a year before he launched his presidential bid? One reason you may not have known was he withdrew before most people had had a chance to notice.

Trump actually withdrew on February 12, and the Marist Poll was conducted on March 6 - three weeks later. Despite Trump no longer being a candidate, Marist Poll (and their media partners) made the “news judgment” that he should still be included. It would be more interesting.

At the same time this poll was conducted, Howie Hawkins was pretty much a lock to win the Green Party nomination for governor. Thanks to his strong showing in the 2010 governor’s race, the Green Party had automatic ballot access in the 2014 race as well.

Hawkins was not included in the poll, and struggled throughout his campaign to get the same media coverage that other parties received. He still went on to get 184,419 votes, or 4.8% - nearly triple his 2010 result.

Meanwhile, back in 2018, Hawkins is being excluded from polls yet again, but Sex & the City star Cynthia Nixon is getting plenty of attention. Despite not having secured a place on the Democratic primary ballot, nor any obvious qualifications to be a governor, she is being included in polls by all three groups we contacted. Television stars sells papers.

After we ran our News Growl special report on this story on Friday we asked Howie Hawkins for his thoughts. Here’s what he said:

The exclusion of third party candidates by polling organizations reflects the broader media bias against third parties. That bias underlines the media’s own preference to focus more on handicapping the candidates than reporting on their policy differences. Ironically, if the news media insisted on polling for support of third party candidates, whose votes clearly affect the chances of major party candidates, they would have more accurate reporting on the horse race.

Hawkins’ last point, that the media should have an interest in ensuring that third party and independent candidates are polled, is possibly a way through this current mess.

The media (and their partners like Marist Poll) are chasing ratings, not engaging in some sort of conspiracy to exclude candidates. If enough voters speak up and tell the media they want more accurate polling and then back up their words by changing the channel or not buying papers, they just might listen.