The runoff election between Commissioner Chloe Eudaly and her challenger, Mingus Mapps, is likely the hottest Portland area election in in November.
Eudaly, no stranger to the challenge of ousting an incumbent, has found herself on the defensive. In the May primary, the vote was split between three contenders: Eudaly with 31.3% of the vote, Mapps with 28.6%, and former mayor Sam Adams, who garnered received 27.7%. Five other candidates divided the remaining 12.4% of the vote. Because no one received over 50% in the May election, there will be a run-off between the top two candidates in November.
Since Eudaly and Mapps received roughly the same share of the vote, and there is no polling data on this race, it’s difficult to say who is in the better position come November. One way to think about the race is to look at Adams’s precinct vote totals and see if these are more likely to shift to Eudaly or to Mapps.
This is what we’ve done. Using precinct-level returns from the each of the 81 precincts, we have explored the political, geospatial, and demographic relationships between the precincts won by Eudaly, Mapps, and Adams.
The TL;DR version is this: Mapps looks to be in a strong position come November. There is a strong relationship between vote shares and the demographic characteristics of the precincts where Mapps and where Adams performed well.
However, this comes with an enormous caveat: turnout in the May primary was 51.55% and is likely to surpass 80% in November. The change in the Portland electorate from a May primary to November general may be the key to a Eudaly victory, if she can garner a strong majority of these voters, meanwhile working to carve off at least some of those who voted for Adams.
Finding 1: Mapps and Adams Competed for the Same Precincts
When this race started, Sam Adams was criticized for running against Chloe Eudaly. However, precinct level returns indicate quite clearly that it was in fact Mingus Mapps who should have been complaining. The precincts in which Mapps did well were also the precincts in which Adams did well, while those precincts favorable to Eudaly tended to be unfavorable to both Mapps and Adams (in the scatter plots below, the size of each dot is proportional to the number of registered voters in the precinct).
From these results, it certainly appears that if Adams had not entered the May primary, Mingus Mapps would be picking out the furniture for his office in City Hall.
Finding 2: Demographics Are Not Destiny, But They Sure Are Interesting
We can better understand the nature of this relationship by looking at the differences and commonalities between the precincts that supported Eudaly and those who supported her opponents. Some of this information – such as average voter age – can be gleaned directly from the public voter files. Other quantities cannot be, but we have used a geospatial methodology to construct precinct level estimates of race, income, and educational attainment.
Unlike the August runoff between Dan Ryan and Loretta Smith that we profiled last week, the geographic pattern here is more complex, as it would be given the dynamics of a three candidate race.
Portland’s ‘Kremlin‘ – the mostly affluent, liberal neighborhoods in inner Portland that have determined many local elections – is split between Eudaly and Mapps, with Mapps performing especially well in wealthy and residential neighborhoods like Laurelhurst and Eastmoreland. This fits well with an early position statement of Mapps against a Eudaly proposal to reform Portland’s neighborhood association system.
Meanwhile, Adams won both the very wealthy neighborhoods in West Portland (where Mapps did well) and some lower-income neighborhoods on the Eastern edge of Southeast Portland (where Eudaly did well).
Although maps can give a good sense of the geographic distribution of support, they are not well equipped to capture the relationship between a candidate’s vote share and a particular precinct-level variable. In the figure, we see that Eudaly has a clear advantage in precincts with a lower avereage age. Why might that be?
Eudaly has made a point of advocating for renters, a disproportionately young group, so it stands to reason that precincts with more young people (therefore more renters) would vote for Eudaly at a higher rate (with appropriate warnings against the ecological fallacy).
Eudaly also performed well in low-income precincts and lagged in most of the high-income precincts. This may have to do with the higher rates of homeownership in these areas, which, coupled with Eudaly’s stance against Portland’s neighborhood associations, translated into reduced support in those areas–shown clearly in the figure.
Precincts with higher median incomes, educational attainment, and older residents tended to support Mapps and Adams. This would seem to provide Mapps a smooth path to victory in November. Again, it’s clear that Adams performed well some lower income precincts – likely good news for Eudaly.
But back to the turnout question. What are the characteristics of the 30% of voters who didn’t vote in May but will vote in November, and do they line up more with Mapps voters or Eudaly voters?
Canyon Foot ’20 and Paul Gronke
The runoff election between Commissioner Chloe Eudaly and her challenger, Mingus Mapps, is likely the hottest Portland area election in in November.
Eudaly, no stranger to the challenge of ousting an incumbent, has found herself on the defensive. In the May primary, the vote was split between three contenders: Eudaly with 31.3% of the vote, Mapps with 28.6%, and former mayor Sam Adams, who garnered received 27.7%. Five other candidates divided the remaining 12.4% of the vote. Because no one received over 50% in the May election, there will be a run-off between the top two candidates in November.
Since Eudaly and Mapps received roughly the same share of the vote, and there is no polling data on this race, it’s difficult to say who is in the better position come November. One way to think about the race is to look at Adams’s precinct vote totals and see if these are more likely to shift to Eudaly or to Mapps.
This is what we’ve done. Using precinct-level returns from the each of the 81 precincts, we have explored the political, geospatial, and demographic relationships between the precincts won by Eudaly, Mapps, and Adams.
The TL;DR version is this: Mapps looks to be in a strong position come November. There is a strong relationship between vote shares and the demographic characteristics of the precincts where Mapps and where Adams performed well.
However, this comes with an enormous caveat: turnout in the May primary was 51.55% and is likely to surpass 80% in November. The change in the Portland electorate from a May primary to November general may be the key to a Eudaly victory, if she can garner a strong majority of these voters, meanwhile working to carve off at least some of those who voted for Adams.
Finding 1: Mapps and Adams Competed for the Same Precincts
When this race started, Sam Adams was criticized for running against Chloe Eudaly. However, precinct level returns indicate quite clearly that it was in fact Mingus Mapps who should have been complaining. The precincts in which Mapps did well were also the precincts in which Adams did well, while those precincts favorable to Eudaly tended to be unfavorable to both Mapps and Adams (in the scatter plots below, the size of each dot is proportional to the number of registered voters in the precinct).
From these results, it certainly appears that if Adams had not entered the May primary, Mingus Mapps would be picking out the furniture for his office in City Hall.
Finding 2: Demographics Are Not Destiny, But They Sure Are Interesting
We can better understand the nature of this relationship by looking at the differences and commonalities between the precincts that supported Eudaly and those who supported her opponents. Some of this information – such as average voter age – can be gleaned directly from the public voter files. Other quantities cannot be, but we have used a geospatial methodology to construct precinct level estimates of race, income, and educational attainment.
Unlike the August runoff between Dan Ryan and Loretta Smith that we profiled last week, the geographic pattern here is more complex, as it would be given the dynamics of a three candidate race.
Portland’s ‘Kremlin‘ – the mostly affluent, liberal neighborhoods in inner Portland that have determined many local elections – is split between Eudaly and Mapps, with Mapps performing especially well in wealthy and residential neighborhoods like Laurelhurst and Eastmoreland. This fits well with an early position statement of Mapps against a Eudaly proposal to reform Portland’s neighborhood association system.
Meanwhile, Adams won both the very wealthy neighborhoods in West Portland (where Mapps did well) and some lower-income neighborhoods on the Eastern edge of Southeast Portland (where Eudaly did well).
Although maps can give a good sense of the geographic distribution of support, they are not well equipped to capture the relationship between a candidate’s vote share and a particular precinct-level variable. In the figure, we see that Eudaly has a clear advantage in precincts with a lower avereage age. Why might that be?
Eudaly has made a point of advocating for renters, a disproportionately young group, so it stands to reason that precincts with more young people (therefore more renters) would vote for Eudaly at a higher rate (with appropriate warnings against the ecological fallacy).
Eudaly also performed well in low-income precincts and lagged in most of the high-income precincts. This may have to do with the higher rates of homeownership in these areas, which, coupled with Eudaly’s stance against Portland’s neighborhood associations, translated into reduced support in those areas–shown clearly in the figure.
Precincts with higher median incomes, educational attainment, and older residents tended to support Mapps and Adams. This would seem to provide Mapps a smooth path to victory in November. Again, it’s clear that Adams performed well some lower income precincts – likely good news for Eudaly.
But back to the turnout question. What are the characteristics of the 30% of voters who didn’t vote in May but will vote in November, and do they line up more with Mapps voters or Eudaly voters?
We’ll save that for an upcoming post.