In House races like NY19, where polling is scarce, any indication of support or new data point can prove compelling. While it’s important to consider all the predictive factors, to avoid missing the forest for the trees, it’s also necessary to highlight polls, arguably the most reliable data points, when they arise.
A new poll from the campaign of Erin Collier, a farmer and agricultural economist from Cooperstown, acquired exclusively by Chronogram, shows that despite her fundraising disadvantages, she may still have a shot at the nomination.
The meat and cheese
The Change Research poll of 319 likely Democratic voters, conducted on June 16th and 17th, at first shows predictable results. Antonio Delgado, a lawyer from Rhinebeck, holds a commanding lead at 21 points, followed by Flynn, a business executive from Greene County, and Pat Ryan, a veteran and businessman from Kingston, both with 14 points. Behind them are Jeff Beals, a diplomat and teacher from Woodstock with 9 points, Collier and Dave Clegg, a lawyer from Kingston, with 5 points, and Gareth Rhodes, a former press aide to Andrew Cuomo, trailing at 4 points. Twenty seven percent of respondents said they were undecided. This is about what would be expected based on fundraising and other data points, though the results for Rhodes and Beals are somewhat shocking.
The initial results match up with other polls of the race, which have shown either a two-way race between Ryan and Delgado or a three-way race between Ryan, Flynn, and Delgado.
When voters “learn about [Collier’s] impressive background,“ along with the backgrounds of some of the other candidates (all positive) she closes the gap. The final tally shows Delgado and Ryan tied at 18, Collier on their heels at 17, Flynn falling behind at 12, Beals at 7, Clegg at 4 and Rhodes at 3. 19 percent said they were still undecided. These results are astonishing to say the least.
Change Research also tested the importance of certain background and policy factors. Three quarters of voters “find [Collier's] longstanding ties to the district to be a convincing reason to vote for her,” while nearly half said they “would be less likely to vote for a candidate who recently moved to the district to run for Congress.” This highlights the importance of community ties, which Collier–an eighth generation Cooperstown resident who boasts Collierstown adopting her family’s eponym–has in abundance. It also showed health care to be of great importance with 28% of voters, a plurality, citing it as the primary issue they want their Representative to focus on in Congress. It also found that more than two thirds of voters favor Medicare-for-All, a progressive solution supported by all NY19 candidates except Ryan and Delgado.
The poll also had some insights on the strength of the #MeToo movement on the role of women in politics. It showed that 82% of voters had a favorable view of New York Senator Kirsten Gillibrand, a scion of the women’s movement in Congress who notably helped trigger the ouster of Senator Al Franken after he was accused of sexual misconduct. Earlier polling by Change Research also showed female candidates in other races picking up a lot of initially undecided voter, with 49% of undecideds breaking for women vs. 38% of decideds. These numbers are somewhat encouraging for Collier and explain her rise after information about candidate is offered. In the final tally, 8% of previously undecideds committed to a certain candidate. It seems Collier picked up many of those.
The nuts and bolts
Looking at the methodology of the poll, its origins, and the reliability of a single poll in predicting race outcomes is key to understanding this particular poll in proper context.
According to FiveThirtyEight’s pollster ratings, Change Research gets a C+ rating (marginally above average) with 57% of races called correctly from a sample of seven polls. For reference, Marist Polling, which get’s an A rating from the reputable polling blog, called 85% of races correctly with a much larger sample size of 159 polls. Change Research is slightly less than respectable, but not so much so that its polls should be dismissed outright. Still, the firm uses online surveys which have proven thus far to be a dubious method of gathering this kind of data compared to traditional live polls. Hence their C+ rating.
A sample size of 319 voters is small, about half the sample size of a normal House primary poll, and polling “likely” voters as opposed to “registered” voters may discount a possible upswing in turnout as part of the well-documented blue wave. Moreover, the poll only introduces new information about four of the candidates–Collier, Delgado, Flynn and Ryan–making it much more difficult to determine whether another candidate’s background might resonate and cause them to surge in much the way Collier does.
As with all single polls, this shouldn't be taken as a perfect predictor of the race's outcome. Moreover, as an internal poll for the Collier campaign, we should assume she is overvalued to some extent. Finally, considering its methodology, the numbers should be taken with a grain of salt.
Finding a signal through the noise
If the numbers show nothing else, it’s that Collier would be well served by getting her message out there and that she could ride the momentum of the #MeToo movement to victory on the 26th. Still, it will be logistically difficult for her campaign to disseminate the kind of information that this poll shows will move voters into her corner considering her campaign cash is the lowest of anyone in the race: just under $100,000 at last count.
Still, her campaign feels confident.
“At every forum and at every door people are excited to vote for Erin,” said Collier spokesman Paul Feingold who revealed a plan to capitalize on this new information. “We'll have volunteers getting the word out across the district this next week.”
“Erin's message is the most compelling in the field and she has a path to victory,” he said, adding, “we're confident in our plan to get Erin's message out.”
While we don’t know if the Collier campaign will succeed, one thing is for sure based on this poll; this is still anyone’s race.
