Four things to know about the upcoming poll of the Philly mayor’s race

If you’re one of the people closely following the Philadelphia mayor’s race, you might have noticed that we don’t have a ton of data about where this race stands, at least compared to a presidential or U.S. Senate race that sees more polling.

But this Friday, the mayor’s race is going to get its first independent, public poll.

The good-government group Committee of Seventy, in partnership with a handful of other organizations, is planning to release a survey of 1,000 Philadelphians that will show where the top candidates stand with just a few weeks to go until the May 16 primary election.

It comes as five top contenders with paths to victory are doing everything they can to separate themselves from one another. Campaigns do their own polling from time-to-time, but they’re unlikely to release it publicly if it doesn’t look good for them. Grocer Jeff Brown’s camp touted an internal poll in February showing him in the lead — and hasn’t published one since.

This is why news organizations like The Inquirer consider independent polling to be the most reliable. Generally, those polls are published no matter what the results show.

» READ MORE: Cut through the noise of the mayor’s race with The 100th Mayor weekly newsletter from The Inquirer

We wanted to explore what can — and can’t — be learned from a single poll, so we chatted with pollsters Berwood A. Yost, director of the Center for Opinion Research at Franklin and Marshall College, and Christopher Borick, director of the Muhlenberg College Institute of Public Opinion.

Here are four things to know ahead of Friday:

Primary polls aren’t nearly as good at predicting winners as general-election polls. Even if the poll is conducted perfectly, voters are still more likely to change their minds ahead of a primary because, unlike in a general election, they aren’t swayed by party affiliation.

There is still a lot of money left to be spent on this race, both by campaigns and by the outside groups supporting them. Voters who are reached by advertisements, door-knockers, and other communicators can and will change their minds as we get closer to Election Day.

Yost said it’s also challenging for pollsters to correctly sample respondents who are representative of the likely electorate because turnout is hard to predict in off-year elections. He said more pollsters used to conduct surveys ahead of mayoral races, but “there were some pretty spectacular failures over the years that caused people to get out.”

“There’s just not much incentive to poll in these primaries because it’s too easy to be wrong,” Yost said.

So let’s say Joe Schmo leads Joe Blow by three points, and the margin of error is plus-or-minus 5 points. This means that the pollster’s best guess is that Schmo is ahead — but there’s some uncertainty there, especially because it falls within the margin of error. Blow could even be ahead. In this example, the candidates could be up to 10 points apart and still be within the margin.

That’s how margins of error work. They tell the person consuming the poll results how certain or uncertain the pollster is. Any poll, Yost said, is still only reflective of a sample of the electorate. The real key is ensuring that sample is representative.

In polling, it’s more instructive to be able to compare polls that use the same methodology but are taken at different times so that you can see how attitudes change over time. Unless the same pollsters conduct another survey closer to Election Day, it’s not likely we’ll have an apples-to-apples comparison.

» READ MORE: Get to know the top candidates for Philly mayor

Yost said that, especially if multiple candidates are polling at about the same levels, the poll will be more useful to identify tiers, not necessarily to place candidates in first, second, third, etc.

“I’m not terribly interested in that horse-race number at this point,” Yost said. “If there are a few people that separate themselves, that’s interesting. But I wouldn’t put too much stock in that. What I’m doing at this point is trying to identify the tiers of candidates.”

Borick said it’s empowering for voters to get some information from independent pollsters as opposed to none. But he said the downside can be if polling moves voters away from candidates they most align with.

“And if that is based on flawed data points, or limited to maybe a poll or two that wasn’t necessarily indicative, I think that’s problematic,” he said.

Yost said the horse-race findings that show where each candidate stands are sometimes the least important finding in a survey like this. He said he’d be more interested in respondents’ answers to questions that probe how likely they are to vote, how engaged they are, or whether they are familiar with the candidates.

The Committee of Seventy said poll respondents will also be able to rank their choices in order to demonstrate what the outcome would be if Pennsylvania implemented a ranked choice voting system. Such a system uses a formula to factor in voters’ second, third, and subsequent choices if one candidate does not win a majority outright.

Given the Democratic Party’s huge voter registration advantage in the city, the winner of the primary is well-positioned to win in November. That, coupled with the large field, could mean that the next mayor will have been nominated by less than 30% of primary voters who went to the polls.

Borick said he “loves” that Seventy is asking respondents to rank their choices.

“It’s neat to think what would it look like,” he said. “And would it be different?”