Your guide on where local polls come from, how to interpret them and which to trust.
by Rachel Kahn, The City
Feb. 20, 2025, 2:05 p.m.
The 2025 mayoral election season is off to a roaring start.
Earlier this month, President Donald Trump’s Department of Justice requested Mayor Eric Adams’ federal corruption charges to be dropped, potentially allowing him to campaign without being on trial at the same time. Meanwhile, lots of challengers have already lined up to run against the mayor in the June Democratic primary. And the spectre of former governor Andrew Cuomo still looms over the fray.
With so much uncertainty hanging over the contest, many New Yorkers may look to political opinion polls to help predict the future. From recent headlines, you might be justified in thinking that our former governor is a frontrunner — without officially joining the race.
But that logic, says director of Emerson College Polling Matt Taglia, is flawed.
“Polls are just a snapshot in time, and there is a margin of error,” Taglia told THE CITY. “We’re giving you a range of what is possible.”
So, how should voters interpret the many local surveys they’ll see in the next few months?
We spoke to a number of polling experts to figure that out:
Where do polls come from?
Political polls usually come from one of a few places: an academic institution, a newspaper, an independent polling body or a political consulting firm. They can also come from within a campaign.
Academic institutions will often partner with news organizations to conduct polls — like this one from Emerson and PIX11 that put Cuomo in the lead with 33% of respondents ranking him as their first choice as of early February. In the region, schools with powerhouse polling centers include Marist University, Emerson and Siena College.
You may also see polls from independent consulting or polling organizations, like this one by Bold Decision released in late January.
“Not all polls are equal,” said Adam Rosenblatt, a partner at Bold Decision. “But we try and model ourselves in the way that the academics do.”
Campaigns also pay tens of thousands of dollars to number-crunching firms to conduct polls on their behalf.
According to George Fontas, the founder of political consulting agency Fontas Advisors, internal polling is more focused on issues and messaging, which allows candidates to hone their campaign strategies. But that data is rarely made public, since candidates don’t want their competitors to know their plans.
If candidates do release polling data, there will always be a strategic reason — likely to counter or confirm an existing narrative.
“No campaign is releasing a poll unless there is a clear benefit to them,” said Fontas.
How do I know which polls to trust?
To assess whether a poll is legitimate, experts say you should look at a few key things:
Sample size
Who was polled
Methodology
Transparency
Funding
Partisan affiliation
Let’s break that down.
An important part of poll methodology is the sample: the people who were surveyed. The standard sample size for a national poll is 1,000, but for New York City, experts say to look for a sample size of at least 700 or 800. The smaller the sample size, the larger the margin of error.
“If you have, say, 400 responses, then the margin of error there is going to be like 5%, regardless of if it’s New York or Omaha,” said Taglia from Emerson. “I wouldn’t discredit anything that has a smaller sample size, but it’s going to be hard to say much about that.”
You should also look at who was polled. Polling outfits do their best to try and get a “representative” sample — but what that means changes from election to election. Some polls are asking registered voters, and some are asking likely voters, and these are both different groups.
“It doesn’t need to be reflective of the city as a whole, because the city as a whole doesn’t vote,” explained Fontas. “So is it reflective of the 20 to 25% of New Yorkers who will vote in a primary?”
Voter turnout in New York City is often just a sliver of the total number of registered voters. For example, the 2021 mayoral primary had the highest turnout rate in decades — but still only 26.5% of registered voters cast ballots.
Pollsters generally choose who to survey based on voter file data, or U.S. Census data. And back when landline phone polling was more common, it was easier to get a representative sample.
“What they would do is something called random digit dial, which means they would basically take a list of randomly generated phone numbers and dial them,” said Taglia. “Back then, the idea at least was that there were enough folks with phones who will pick up the phone and answer a poll — that a random sample would produce a representative sample of whatever population you’re studying.”
But that is no longer the case. Now, pollsters use varying methods to reach people, including phone calls, texts, online panels, or other online surveys — sometimes all at once.
Ideally, if you’re reading a poll, you want to check how they surveyed people, the demographics of who was surveyed, what questions they asked and what the margin of error was.
“If someone is just saying ‘I have a poll that shows X’ and they’re not releasing numbers, they’re not releasing details,” said Rosenblatt. “If they’re not sharing that type of stuff, it gets questionable.”
Another thing to look for when reading a poll is: Who paid for it? Polls are expensive, so making sure you know who sponsored it — i.e., who paid the bill — is important, experts say.
For example, though Rosenblatt’s poll published their methodology, its funding source is not disclosed. Rosenblatt told THE CITY that the funder “wishes to remain anonymous,” but is not affiliated with any mayoral campaign.
The influence of money is something you have to worry about less with academic polls because they do their research independently and without sponsorship. But with other for-hire polling bodies, always look for the fine print.
“Was that organization aligned with any particular campaign at the time when they released it? Are they on the payroll of the Democratic Party, the Republican Party? Or are they doing this completely independent of any potential client?” said Fontas, who has done both independent polls and polls for campaigns.
If you can’t tell whether a poll is politically biased, even if its methodology is sound, take it with a grain of salt — especially if the numbers are coming from inside a campaign.
When in doubt, Lee Miringoff, the senior director of the Marist Institute for Public Opinion, recommends people look at the national political analysis website fivethirtyeight.com, which has a ranking tool of different polling groups based on their track records and methodological transparency.
So my poll is trustworthy — does that mean it can tell me who will win?
Even the best polls are not predictive. Polling represents a “snapshot” of one moment in time — and as of right now, that moment is too early to show much.
Miringoff emphasizes that the mayoral polling we’re seeing now isn’t telling us much that we don’t already know.
“Polls now will just tell you that Adams is not that popular, but that there’s not necessarily someone who’s distancing him in a convincing way,” said Miringoff.
He’s seen the Cuomo numbers, but feels that being “way ahead” at 35% isn’t indicative of where the race will go. “He has the name recognition, that’s what generates that,” he said of the apparent lead.
A lot can change over the course of the race. Miringoff points to polling in New Hampshire in 2015, one year before the Republican presidential primary of 2016.
“Trump had 1%,” he said. “It’s very time-sensitive.”
Polls also can’t predict how campaigns will impact voters — significant ads, speeches, or released plans could change someone’s mind in a way they haven’t anticipated — or how the field of candidates will develop.
“There are major political storylines that are playing out even as of today surrounding the mayor,” said Fontas.
So, with so much unknown, what’s the point of doing polling this early?
For candidates, Fontas says that polling can help shape perception, which can be important for their campaign strategy.
“Early polling can help shape the early stages of the race, which can affect one’s ability to fundraise, one’s ability to get additional earned media,” he said.
Will we see any polling for the city’s other offices up for election?
Even though New Yorkers will likely be bombarded with mayoral polls, you’re probably not going to see much polling on other races, like comptroller, or for candidates running for City Council.
For one, it’s harder to get useful data on these races — especially this early in the cycle — because the candidates have less name recognition, experts said.
“In smaller races, it becomes more problematic because the candidates aren’t that well known,” said Miringoff. “Talking about comptroller or public advocate — even though those are citywide [races], the numbers on those aren’t going to be particularly dependable.”
It’s also harder to sample the right people: calling people up based on their zipcode won’t align with their City Council district, and you still need to have a large enough sample size (around 800) even though you’re working with a smaller overall population.
Plus, as we mentioned earlier, polling is expensive. Another Emerson pollster, Spencer Kimball, previously told THE CITY that because those smaller races don’t get as much news coverage, there’s less publicity in it for the polling groups.
“There’s value to that for the institutions. When you start doing some down-ballot races that people aren’t following, you’re not going to get that type of coverage. And so it loses value,” said Kimball.
As we get closer to the election, you can expect to see a lot more polls — especially as the playing field of candidates becomes clearer. (Remember, we won’t know who is officially on the ballot until early May.)
According to Miringoff, Marist is waiting a bit before jumping into the fray.
“I would hate to be in the field right now, with the Trump and Adams thing and all that,” he said.
If big news breaks — like when the DOJ announced it planned to drop the federal corruption charges against the mayor — it can skew the results of a poll that is only partially complete.
“You could’ve gotten half a poll done one way and half done the other way,” Miringoff said.
Comments