July 2024 Mayoral Poll Results

Last Updated: July 24, 2024
The Mayoral RaceThe Mayoral Race

Who would win if the election were held today?

The race is a statistical tie with no clear winner. Using current polling, London Breed wins with just 50.99% in the final round of ranked choice voting (RCV)—but this victory is within the 3.1% margin of error and ignores the 18% of voters who are still undecided.

Compared to May's mayoral poll that showed 28% undecided, this month's poll showing 18% undecided indicates that voters are starting to line up behind candidates.

Daniel Lurie is consolidating 2nd rank votes, growing his share from just 14% in May, to 22% today. The next-biggest change is the surge in 1st rank votes for Mark Farrell - growing from just 16% in May to 21% in July. Mayor Breed has mostly stayed put, with her first rank votes increasing within the margin of error from 21% to 24%, and her second rank votes just edging past the margin of error to grow from 6% to 10%.

San Francisco's ranked choice voting system means voters can rank more than one candidate in order of preference. To simulate this, we asked voters who they would vote for if the election were held today, and who they would rank in second, third, and fourth place.

About the poll: GrowSF commissioned FM3 Research to run a citywide poll of 1,066 San Francisco voters from July 9 to July 18, 2024. The results have a ±3.1% margin of error at the 95% confidence level.

Our poll question

The question: Thinking about November's mayoral election, who would be your first choice for Mayor if the candidates were as follows:
  • London Breed, San Francisco Mayor
  • Ahsha Safaí, San Francisco Supervisor
  • Aaron Peskin, San Francisco Supervisor
  • Daniel Lurie, Nonprofit Executive
  • Mark Farrell, Small Business Owner
  • Someone else (Other)

To learn second, third, and fourth place rankings we asked follow-up questions "Who, if anyone, is your second/third/fourth choice for Mayor?". If the respondent indicated they were undecided or unsure, we asked, "Would you lean toward one of the other candidates or would you have no second choice?". If they responded that they'd have no second choice, we ended the question and moved on.

Candidate names were presented in random order to each respondent to avoid order bias.

How each round of voting plays out

GrowSF built a ranked choice voting simulator to model the election. The simulator uses the poll data to simulate how votes transfer from one candidate to another in each round of RCV.

Note: "Exhausted" votes are ballots where all the ranked candidates were eliminated. These ballots won't be counted in subsequent rounds.

Round 1

Candidate% of Active Ballots% of All Ballots
Breed30%24%
Farrell26%21%
Lurie20%16%
Peskin16%14%
Safai8%7%
Undecided or exhausted18%

Safai is eliminated because they had the fewest votes. Here's how Safai's votes transfer for the next round:

FromToPercent
Safai
Breed29%
Farrell13%
Lurie10%
Peskin28%
Exhausted21%

Round 2

Candidate% of Active Ballots% of All Ballots
Breed33%26%
Farrell27%22%
Lurie21%17%
Peskin19%15%
Undecided or exhausted19%

Peskin is eliminated because they had the fewest votes. Here's how Peskin's votes transfer for the next round:

FromToPercent
Peskin
Breed30%
Farrell14%
Lurie20%
Exhausted37%

Round 3

Candidate% of Active Ballots% of All Ballots
Breed41%31%
Farrell32%24%
Lurie26%20%
Undecided or exhausted25%

Lurie is eliminated because they had the fewest votes. Here's how Lurie's votes transfer for the next round:

FromToPercent
Lurie
Breed20%
Farrell47%
Exhausted34%

Round 4

Candidate% of Active Ballots% of All Ballots
Breed51%35%
Farrell49%33%
Undecided or exhausted32%

Breed wins!

How do votes transfer in the first round?

Second Choice
BreedFarrellLuriePeskinSafaiOtherNo 2nd Choice
First Choice
Breed-19%25%15%9%0% 31%
Farrell10%-71%5%5%1% 9%
Lurie18%46%-7%11%2% 16%
Peskin20%4%7%-48%1% 18%
Safai29%13%10%29%-0% 18%
Other7%0%0%0%14%- 79%

This table shows how votes transfer from one candidate to another.

For example, look at Mark Farrell's row. 10% of Farrell's voters rank London Breed as second choice, an astonishing 71% rank Daniel Lurie, 5% rank Aaron Peskin, 5% rank Ahsha Safai, and 9% don't make a second choice at all.

The other big stand-out number is that 31% of London Breed voters only vote for her and don't rank any other candidates (this is called "bullet voting"). This might be because the race is so far out and the other candidates' name recognition is low, but if it carries forward to November, this high number of bullet votes would mean these voters would lose out on a chance to have their voices heard should Breed not make it to the top two in the final round.

The GrowSF TakeArrow Decorator

This race is a statistical dead heat. In our simulations, small changes swing the election in favor of any of the top three candidates. This means #2 votes will play a key role in who becomes mayor. For example, if more Peskin voters rank Lurie #2, he could edge out Mark Farrell in the 3rd round of RCV and clench victory from Mayor Breed. Or if a few more Safai voters rank Mark Farrell second, he can sail to victory. Mayor Breed can also easily win if she expands her 2nd-rank appeal to Lurie, Safai, or Peskin voters.

Compared to our May 2024 poll, the number of undecided voters has decreased from 28% to just 18%, indicating that the candidates are starting to make an impression on the electorate.

Mayor Breed's first-rank votes have increased slightly, from 21% to 24%, while Mark Farrell has gained the most ground in first-rank votes, growing from 16% to 21%. Daniel Lurie's first-rank votes stayed steady at 16%, but Lurie has significantly increased his share of #2 votes, growing from 14% to 22%. Peskin remains in fourth place, with his first-rank votes staying within the margin of error at 13%.

34% of our respondents didn’t list a second choice candidate, and 52% didn’t list a third. Overall, 32% of voters missed the chance to have their preferences reflected in the final results because their preferred candidates lost without ranking anyone who made it to the final round of RCV.

Candidate favorabilityCandidate favorability
The question: Is your impression of each candidate generally favorable or unfavorable?
Mayor London Breed
8%
28%
24%
31%
San Francisco Supervisor Aaron Peskin
7%
23%
11%
33%
Mark Farrell
17%
28%
12%
12%
Daniel Lurie
11%
37%
14%
8%
San Francisco Supervisor Ahsha Safai
4%
21%
18%
15%
Very favorable
Somewhat favorable
Somewhat unfavorable
Very unfavorable
The GrowSF TakeArrow Decorator

Mayor Breed’s favorability numbers increased slightly compared to our May 2024 poll, with her unfavorables dipping slightly in turn. Mark Farrell and Daniel Lurie both picked up huge gains in favorability, with Farrel mostly increasing his "strong favorable" and Lurie dominating the "somewhat favorable" rankings. Aaron Peskin also improved his favorability, but at the cost of growing "strongly unfavorable" opinions.

Sign up so you don't miss the next GrowSF Pulse!

Methodology

fm3
Dates
July 9 - July 18, 2024
Survey Type
Dual-mode voter survey
Research population
San Francisco voters
Total interviews
1,066
Margin of sampling error
±3.1% at the 95% confidence level
Contact methods
Telephone calls
Email invitations
Text invitations
Data collection modes
Telephone interviews
Online interviews
Languages
English and Chinese

Footer

Sign up for GrowSF's weekly roundup of important SF news

Sign up for our weekly newsletter

Follow us on social media

TwitterFacebookInstagram
Line art illustration of San Francisco