Warning: This post is rather long - sorry for if it comes across as rambling - just trying to be very clear when explaining specifics. There's a TL;DR at the end. (Flared as discussion because this is a discussion post that at least kinda reads like a textbook question.)
Hey y'all, I've started to dive into precinct level results for AZ (although stuff isn't finalized still) in order to find potentially notable results that could serve as useful messaging knowledge in a council race - and I noticed an interesting trend at our local level that I was wondering if I could get some help deciphering.
So this year our fine city voted on 5 funding bonds - Sewers, Storm Drains, Public Safety (Cops+Fire), Parks, and "Municipal Infrastructure", which were all sold by the council (which wrote a bipartisan statement in favor of it in our election info pamphlet) as "vote for this so normal city government functions continue". There was no real organized opposition, even against the cop bond. These same bond topics were voted on in 2016, and the %Yes on them each increased in 2020 from 2016 by 4-5% (although the cop bond only slightly ticked upwards), and roughly a similar gain from the average %Yes on bonds in 2012 (which granted were for different topics) to the 2016 results.
The 5 measures this year passed with 79%, 78%, 70%, 76%, and 68% respectively.
Before I continue, here's a very quick primer on Tempe's political geography for necessary context, with 2016-Pres, 2018-Sen, 2018-AG, and 2018-Gov results listed (barometers for national, moderate statewide, generic D statewide, and progressive statewide sentiments), as well as a rough guess on Biden's % this year (
not margin).
On this map of city zones the colors correspond to the following regions:
- Blue: Rich, very white (72%) suburban area. Large yards, almost universally multi-story non-rented buildings, interspersed horse ranches. Old-money area vs. new-money developments to the west in Ahwatukee. Relatively conservative city council voting trends. Clinton+12.3, Sinema+18.6, Contreras+7, Garcia+0.1 - Garcia in particular won by only 12 votes. 2016 3rd party vote of 6.7%. ~58.9% Biden'20.
- Brown: Very white (79%) retirement community surrounded by single-story middle-class residences (see Yellow), with the retirement community casting the deciding majority of the votes in the district. Relatively conservative city council voting trends. Clinton+10.4, Sinema+30.3, Contreras+13.6, Garcia+0.2 - Garcia only won by 2 votes. 2016 3rd party vote of 9.2%. 64.82% Biden'20 (since this is just one precinct, this is an exact#).
- Red: Predominantly Hispanic (with noticeable Black, Asian, and Indigenous populations as well) low/low-middle income precincts. Predominantly renters. Center-left city council-voting trends with strong spikes in support for Hispanic candidates. 39.2% white. Single-story homes. Clinton+34.8, Sinema+48.1, Contreras+42.3, Garcia+33.8. 2016 3rd party vote of 11.7%. ~71% Biden'20.
- Teal: Noticeable Hispanic minorities (30.5%), albeit not outright minority-majority (vs. red). Low-middle/middle income. Predominantly single-story with interspersed apartments. Roughly 50/50 on rent vs. owned housing units. Relatively center-left voting trends with slight tilt to Hispanics candidates. Clinton+26.9, Sinema+39.4, Contreras+31.3, Garcia+20.9. 2016 3rd party vote of 10.5%. ~66% Biden'20.
- Yellow: White (64.9%), middle class family-with-children residences of varying story sizes in the northern half of the city (Baseline Rd. is used as an N/S dividing line - or sometimes that's US Route 60 - when people rant about their half not getting enough representation on the Council). Slightly more owned than rented residences. Relatively center-left city council voting trends. Clinton+25.3, Sinema+35.7, Contreras+28.5, Garcia+21.5. 2016 3rd party vote of 10.3%. ~63% Biden'20.
- Pink: Slightly whiter (67.8%), slightly richer, less rented, and more conservative clone of Yellow. Almost a sliding scale driving south from Green through Yellow and Pink to Blue. Slightly left-leaning city council voting trends. Clinton+17.4, Sinema+26.8, Contreras+15.3, Garcia+8.3. 2016 3rd party vote of 9%. ~61.5% Biden'20.
- Green: Area around Arizona State University, predominantly students and recent graduates in apartments/condos. Diverse (53.2%, with Asians and Hispanics roughly tied around 20%). Pockets of low/low-middle income Hispanics single-story neighborhoods in the eastern portion of the Green area. Heavy renter population. Relatively liberal city council voting trends. Clinton+25.3, Sinema+35.7, Contreras+28.5, Garcia+21.5. 2016 3rd party vote of 10.3%. ~71.5% Biden'20.
- Purple: Scattered student apartments extending out from the city's urban core (on the purple/green border) mixed in with predominantly single-story residences or condos of middle-income young professionals. Relatively diverse (57.9% white, 20.9% Hispanic, 10.9% Asian), heavy rental population. Strong support for progressive candidates in Democratic primaries. Once had a budding anarchist scene 15-25 years ago. Relatively liberal/meme candidate city council voting trends. Clinton+47.3, Sinema+60.3, Contreras+54.8, Garcia+49.3. 2016 3rd party vote of 10.9%. 73.47% Biden'20 (also an exact# since it's only two precincts).
I decided to map out the results, and since I'm not the biggest QGIS whiz (and getting another Tableau free trial after exhausting my student one was a pain) I instead decided to color in the top 5 precincts by Yes% (blue) and No% (green) for each bond in Dave's Redistricting, with remaining city precincts colored purple. I've circled the area that got my attention.
Bond 1 - Sewers Bond 2 - Storm Drains Bond 3 - Public Safety Bond 4 - Parks Bond 5 - Municipal Infrastructure A lot of these results are what you'd expect:
- Precincts in Red/Teal zones pop up as top No voters on several bonds, as they did when the city voted in 2016. And I can see why lower-income areas may vote against a municipal bond, either out of financial inabilities to pay it or a "please pay more attention to us" sort of protest vote (notice as well that Red leads the city in 2016 %3rd party votes, and not areas like Green or Purple).
- The rich in zone Blue love cops and hate parks, and in general aren't the most fiscally liberal type (so some precincts crop up in the most-often No's on the non-cop/non-park bonds).
- The olds in Brown are more liberal than their counterparts in Blue and thus support Bonds 1, 2 and 5 with the same vigor as Purple/Green, but lead the city in their cop support% and display the same in park support as Blue (albeit not enough to crack the Bottom 5).
- Purple/Green come in hard for everything except the cop bond, which they understandably oppose.
- Liberal neighborhoods in Yellow sometimes punch above their weight and crack the Top 5 on non-cop bonds.
(All these hold up - relatively speaking - when looking at bond results in other parts of the county.)
And then there's the area circled in red, fittingly a precinct named "Redden" by our County Recorder's Office. It's in the Pink zone, and in general this year behaved like the rest of the zone - a roughly 60D/40R vote split across the board. The only result of note was a potentially racially-motivated ticket split in a school board race, where the precinct (as well as the school district as a whole) voted in 2 Democratic white women and 1 hard-GOP white woman, with a Democratic Afro-Indigenous woman narrowly missing out on the 3rd seat (there were some nasty FB posts floating around in this race, and with this deviation from partisanship being the more prevalent in whiter precincts the conclusions are...depressing).
For further stats, it's the northern half of this census tract. ~2.8K people total. And then despite an otherwise milquetoast voting profile this year, it
led the city in No% in 4 of 5 bonds, and was in the top 4 for the last one (the cop bond). In each of the non-cop questions, Redden's No% was anywhere from 0.75 to 2 points higher than the next precinct. And on the cop question, Redden acted less like a precinct that'd split a school board ticket on racial ground and more like Purple/Green zones, only narrowly getting beaten out by far more leftie precincts in the northern part of the city.
I've tried to figure out what the hell that could possibly indicate, but can't find anything else pointing to oddities in that precinct. Their average %No on judicial retention (something you'd expect would be another vehicle to protest vote random shit) is right in line with the county average, and while they were in the upper half of %No on these bonds in 2016, they were far from the top - and had a visible spike in support for cops that's completely gone this year. Presidential third party vote in the precinct is only around 2.89% - higher than state average but in line with neighboring precincts.
I have no clue what could have driven them to change their views on protest voting this year so much - but only at the municipal level - but I damn want to find out, because this is chock full of swing voters at the council level come 2022.
TL; DR: What, in your experience and conversations with people (or as voters yourselves) could explain a sudden increase in voting "No" on city bonds in a relatively liberal and middle class suburban neighborhood, without any other bizarre voting results in other races, or comparable trends in directly neighboring areas? 
