Oregon: PacNW Domination Thread Part II

[Alcon]

Because I was tired of clogging up early voting.  General Oregon thread!

Let's start where I veered off-course:

OREGON -- While this won't hold up or anything, the breakdown of returned ballots so far says a lot.



(edit: flipped benton and polk. grahh)

So far, Democrats have 62% of the two-party ballot returns in Oregon.  In 2004, the final number was 52%.

Some pretty impressive numbers here -- 68% in Benton (Corvallis), 59% in Clackamas (southern Portland suburbs/Bush county), 54% in Jackson (Medford), 80% in Multnomah (Portland), 62% in Washington (west 'burbs).

Again:  Some ancestral voting in the East, and the gap should go down, but still, shows where the enthusiasm is at.  GOP turnout in Oregon is almost universally higher.

Curiosity calculation:  Using SUSA's partisan break-downs to extrapolate this ridiculously, Obama leads turned-in ballots 60%-37%, and gives the following patently ridiculous map.



(edit: flipped douglas and lane.  supergrahh.)

For Democratic salivary glands only.  Even if Obama won by 23 points (lol) it would look nothing like this, with heavier ditro to Multnoco, where indys outnumber Republicans.

Alcon, thanks gain for an awesome contribution.

This map is interesting, although as we both know Obama will most likely only win Oregon by 13-15 points.

In that event, I think that the colors of the top map look correct except McCain will win Linn and maybe Coos and Yamhill counties (I am operating on the assumption that Douglas and Lane are flipped in your initial map). I think that there is a decent chance that Obama will carry Curry (older, wealthier California residents), and possibly Union and Crook Counties. Josephine is still a long shot, although long-term demographic changes favor the Democrats. Jefferson county should go Obama with this type of margin.

[Alcon]

Alcon, thanks gain for an awesome contribution.

This map is interesting, although as we both know Obama will most likely only win Oregon by 13-15 points.

In that event, I think that the colors of the top map look correct except McCain will win Linn and maybe Coos and Yamhill counties (I am operating on the assumption that Douglas and Lane are flipped in your initial map). I think that there is a decent chance that Obama will carry Curry (older, wealthier California residents), and possibly Union and Crook Counties. Josephine is still a long shot, although long-term demographic changes favor the Democrats. Jefferson county should go Obama with this type of margin.

I did a primary-based model.  The results are here.  There's some, um, oddities such as Obama falling in Hood River.  That seems unlikely.  I forget what was close.  McCain won Curry by double-digits in my model, but I dunno.  Deschutes was quite close.  Union and Crook are tough nuts to crack.  I'll definitely be watching you guys very closely on election night.  I would never, ever have seen this race been approaching +15 in Oregon.  Obama has gotten a pretty incredible chunk of persuadables in the state.

Pretty much every model I run has Jackson flipping, which in itself is impressive.  All have Clackamas, Tillamook and Wasco.  Coos is in most, as are Marion.  A sizable minority have Linn, Polk and Yamhill.  Blanking on any others atm...

I don't think any had Jefferson, though, but Madras is so beyond my knowledge, and the scope of the models.  It's a confusing place for me (Crook County too)

No question Obama wins Hood River by at least 10 points....

Clackamas, Tillamook, Marion, and Wasco, and Polk are gone without question, although McCain should fall within a 5-8 point margin.

I feel more concerned about Jackson and Coos counties, although demographic changes in the Medford area tend to favor Dems at the national level. Coos is one of those blue-collar historic timber Dem counties (I believe they went for Mondale in '84) where there might be a bit of reversal from the past few cycles where there was a backlash against the national ticket for perceived excesses on national resource issues and guns. Curry county is a slight long shot, but Southern Oregon is changing, and the socially liberal retiree vote may well swing towards Obama.

Eastern Oregon is interesting, and I have gone out on a limb before predicting a few of these counties. Deschutes is extremely wealthy with many California retirees, but don't forget that this is a county that went with the change message of Lonsdale in the Senate race in '90. Jefferson county has a large Latino population, a Native American population, and wealthy Anglo tourist sector. Union county gave 47% to Dukakis in '88, and I suspect with the student vote that Obama may well carry this county.

[Alcon]

I hastily re-made my Oregon model so I could get more exacted details.  This is the most conservative form.  My other models are sometimes "hot spot models" -- that is, they're meant to exaggerate certain effects to see which counties would swing more under what circumstances.

This is a more standard model, based primarily on 2008 Primary results.  It's the closest I could remember to what I used for the map provided.  This is Obama +14.  Dem margin swing in parentheses.

Baker: McCain +21 (+19)
Benton: Obama +29 (+11)
Clackamas: Obama +10 (+11)
Clatsop: Obama +19 (+9)
Columbia: Obama +19 (+16)
Coos: Obama +5 (+16)
Crook: McCain +21 (+17)
Curry: McCain +12 (+5)
Deschutes: McCain +3 (+11)
Douglas: McCain +17 (+15)
Gilliam: McCain +16 (+18)
Grant: McCain +30 (+29)
Harney: McCain +38 (+15)
Hood River: Obama +14 (-1)
Jackson: Obama +3 (+15)
Jefferson: McCain +12 (+7)
Josephine: McCain +13 (+13)
Klamath: McCain +30 (+16)
Lake: McCain +39 (+18)
Lane: Obama +24 (+6)
Lincoln: Obama +17 (+2)
Linn: McCain +3 (+19)
Malheur: McCain +36 (+15)
Marion: McCain +2 (+7)
Morrow: McCain +16 (+17)
Multnomah: Obama +51 (+7)
Polk: McCain +4 (+7)
Sherman: McCain +26 (+1)
Tillamook: Obama +7 (+9)
Umatilla: McCain +16 (+15)
Union: McCain +18 (+15)
Wallowa: McCain +25 (+16)
Wasco: Obama +6 (+9)
Washington: Obama +16 (+10)
Wheeler: McCain +19 (+23)
Yamhill: McCain +4 (+11)
OREGON: Obama +14 (+10)

Comments welcome -- probably a few oddities (Hood River is real screwy), but eh.  It was the most realistic result I could get through objective means. 

Note: Despite my best efforts, the model punishes places heavy on Obama-voting independents (and lapsed Republicans.)  This probably damages counties like Washington and Hood River, and oddly should damage Multnomah, but it doesn't show it.  Virtually every model I've run has Obama excelling in Multnomah; he did very well in the primary there in every respect.  It also helps that very low GOP registration punishes McCain.

[NOVA Green]

Thanks for the PM! I am a little reluctant to make exact predictions considering that I was a bit conservative in my model during the primaries, but here goes....

Crook county went 47% for Dukakis, and there is a decent population now that commutes to metro Bend, as well as a tourist base that tends to favor Obama.... I consider this county as  a likely flip with Obama +14-15.

Curry county, like Deschutes (Bend) is highly variable depending on the leanings of wealthy California independent retirees... potential Obama narrow flip.

Deschutes county... I really think this might be the first Presidential election where a Democrat might have a chance of winning this socially liberal, fiscally conservative, out of state retiree region. Maybe more of a hunch than anything else.

Douglas county will probably give at least 43-44% to Obama, although nowhere near as close as Dukakis got in'88, the days of the downstate timber voting bloc at the Presidential level is over.

Jackson, I agree is a likely narrow flip for Obama.

Jefferson is an area highly dependent on tourism with a large NA and Latino population which has historically done well for Democrats... this county will flip in a 15 point election.

Linn... this county is right next door from mine, and despite decent Obama numbers in Albany, there are simply way to many voters in the small town and rural areas that identify with the Republican message for Obama to carry this county.

Marion... has historically been a slight Republican leaning county, but the city of Salem has been trending Democratic the past few cycles... Rural parts of the county are not sufficient for McCain to eke out a win.

Multnomah... will go heavily Dem as always, but I think Obama's ceiling is 75-77% here, which would a record number for a Dem. This explains some of the extra Democratic votes in other counties I have mentioned.

Polk Co... between WOSU and the Salem exurbs Obama should get at least 48% here... I personally think that with current trends and the Latino population that Obama will most likely narrowly beat McCain here.

Tillamook will be close, your 7 point margin might be slightly exaggerated or might well be dead on... too close to call.

Union Co will be much closer than your model anticipates... La Grande is the most liberal city in Eastern Oregon (unless you include Bend) and the university vote here should help. Don't forget this is a county that almost went for Dukakis in '88.

Yamhill... probably about right, but the exurban Portland population will keep it a marginal McCain county, unlike Polk which might well flip.

I have other observations, but I don't want to nit-pick each county... don't forget I got slammed in the primaries for underestimating Obama support, particularly downstate.

[Alcon]

The Multnomah number seems a little crazy to me, too.  If only you guys had an open primary, this would be leagues easier.  I agree with nearly everything you said.  I'm still not 100% sold on the competitiveness of some of the counties you mentioned, but with some Multco support leaking -- I can see it.

I also think my model overballed a few other counties (Benton, Clackamas maybe, Linn almost certainly).  And then there are all of those darn ancetral registrations in the East, but they're limited to only a few counties.

My dream model would:

1. Further underball Obama slightly in counties with low Dem turnout vs. GOP turnout in 2004 (generally a good indicator of ancestral registrations)

2. Handle independent voters who couldn't vote in the primary...somehow.

3. Make some kind of regression involving Clinton voters, because they aren't gonna vote the same way in Multnomah as in Morrow

4. Maybe figure out a way to deal with places like Washington, where the ancestral registrations are starting to go the other way

5. Maybe do something about the laundry I skipped in order to write these models ()

[NOVA Green]

Assuming MultCo is 75-25 or 77-23 (rather than the 80% model with is way too high), and Obama wins statewide 13-15%, then I would say that Marion, Polk,  and Jefferson flip.

With regards to the other counties: there is a very good chance that Deschutes, Crook and maybe Curry and Union flip.

Benton county will most likely go Obama +25, although don't forget that West Albany will keep county margins down, plus the fact that the vaunted youth turnout hasn't yet materialized anywhere nationally.

I'm not convinced that you are lowballing Linn county... this is a tough nut to crack in the extremely polarized Oregon electorate.

Your model has much to go for it... although the points you make in your "dream model" are pretty much all valid. Not sure about the laundry situation however

[Alcon]

Hm, well, bump.

Overall, the model fared OK.  The biggest misses were pretty obvious weird results -- parts of Eastern Oregon, Hood River County.  Overall, it seemed to over-estimate Obama among downscale, culturally conservative oldschool Democrats and under-estimate him in the suburbs.  Then again, isn't that pretty much what happened everywhere?

(I knew Hood River looked ridiculous.  lol.)

Table


Map


-------------------------------------------------------------------------------------------------

So far, the only precinct results I have are preliminary ones for Polk County.  Since I'm bored, I'm posting them:

102 Ballston & Spring Valley: McCain +12 (was Bush +20) +8%
106 Perrydale & Buell:  McCain +12 (was Bush +23) +11%
110 Grand Ronde & Willamina: Obama +7 (was Bush +1) +8%
118 Rural Salem & Monmouth: McCain +8 (was Bush +15) +7%
122 Rural Independence: McCain +21 (was Bush +19) +2%
134 Salem City: Obama +1 (was Bush +10) +11%
144 Salem City: Obama +7 (was Bush +5) +12%
146 Salem City: Obama +5 (was Bush +7) +12%
152 Rural Dallas: McCain +17 (was Bush +25) +8%
154 Dallas City: McCain +7 (was Bush +17) +10%
162 Dallas City: McCain +5 (was Bush +19) +14%
166 Falls City & Surrounding: Obama +11 (was Kerry +8) +3%
178 Monmouth City: Obama +12 (was Kerry +2) +10%
182 Independence City: Obama +12 (was Bush +2) +14%

That's pretty boring from what I see, but I figure it might interest you as a local   Maybe you can gleam something out of Polk I couldn't.  (Other than the apparent lack of WOU electoral power.  Independence had a bigger swing than Monmouth?)

[Sbane]

It is interesting that your model was pretty much dead on in Curry county even though it overestimated Obama among the white working class. Curry county first caught my eye due to its small swing towards Obama. Why would all the other counties around it swing more than it?

[NOVA Green]

It is interesting that your model was pretty much dead on in Curry county even though it overestimated Obama among the white working class. Curry county first caught my eye due to its small swing towards Obama. Why would all the other counties around it swing more than it?

Curry county has a large number of wealthy California retirees, whom I thought would swing to Obama by a greater margin. Note that Obama almost won Deschutes county that has a similar high proportion of California retirees (although lower Senior population).

Not sure about the White working class reference.

[NOVA Green]

Hm, well, bump.

Overall, the model fared OK.  The biggest misses were pretty obvious weird results -- parts of Eastern Oregon, Hood River County.  Overall, it seemed to over-estimate Obama among downscale, culturally conservative oldschool Democrats and under-estimate him in the suburbs.  Then again, isn't that pretty much what happened everywhere?

(I knew Hood River looked ridiculous.  lol.)

Table


Map


-------------------------------------------------------------------------------------------------

So far, the only precinct results I have are preliminary ones for Polk County.  Since I'm bored, I'm posting them:

102 Ballston & Spring Valley: McCain +12 (was Bush +20) +8%
106 Perrydale & Buell:  McCain +12 (was Bush +23) +11%
110 Grand Ronde & Willamina: Obama +7 (was Bush +1) +8%
118 Rural Salem & Monmouth: McCain +8 (was Bush +15) +7%
122 Rural Independence: McCain +21 (was Bush +19) +2%
134 Salem City: Obama +1 (was Bush +10) +11%
144 Salem City: Obama +7 (was Bush +5) +12%
146 Salem City: Obama +5 (was Bush +7) +12%
152 Rural Dallas: McCain +17 (was Bush +25) +8%
154 Dallas City: McCain +7 (was Bush +17) +10%
162 Dallas City: McCain +5 (was Bush +19) +14%
166 Falls City & Surrounding: Obama +11 (was Kerry +8) +3%
178 Monmouth City: Obama +12 (was Kerry +2) +10%
182 Independence City: Obama +12 (was Bush +2) +14%

That's pretty boring from what I see, but I figure it might interest you as a local   Maybe you can gleam something out of Polk I couldn't.  (Other than the apparent lack of WOU electoral power.  Independence had a bigger swing than Monmouth?)

Thanks Alcon...

I overestimated Obama performance in Eastern Oregon based on both primary performance, plus historically how Union, Crook, and Jefferson counties have performed with Liberal Democrats.

BTW: Your model was almost dead-on in metro Portland, slightly high in Clackamas, but slightly over in Portland.... 77% Multnomah? WTF? Washington 60%?

Thx for the Polk Co precinct reports, much appreciated!!!! Did you get them from the SoS or county election division?

[Alcon]

I actually underestimated Obama in Multnomah and Washington, and nailed Clackamas (which surprised me)

I just got the Polk results from the web site.  It's a bit too early for me to begin to harass county clerks.

[Platypus]

Clatsop?

[Alcon]

Clatsop?

...?

[Platypus]

(Do you have details for)

[Alcon]

(Do you have details for)

Polk's all I've got.  It may be a week or two before counties release precinct information.  They may have a scattering of votes to add on to the final.

Will post 'em when I get 'em

[Sbane]

It is interesting that your model was pretty much dead on in Curry county even though it overestimated Obama among the white working class. Curry county first caught my eye due to its small swing towards Obama. Why would all the other counties around it swing more than it?

Curry county has a large number of wealthy California retirees, whom I thought would swing to Obama by a greater margin. Note that Obama almost won Deschutes county that has a similar high proportion of California retirees (although lower Senior population).

Not sure about the White working class reference.

Hmm but like you said there are lots of retirees in Medford and Bend, yet those areas swung more than Curry. I guess it could just be an influx of only people from conservative areas or something weird like that, although I doubt it. Is Curry county more expensive than Medford and Bend? That might cause only wealthy retirees choosing that area.

[Alcon]

(Do you have details for)

Clatsop:

Astoria: 65-32 Obama (was 62-36)
Cannon Beach: 71-28 Obama (was 64-35)
Gearhart: 61-38 Obama (was 58-41)
Seaside: 61-36 Obama (was 57-41)
Warrenton: 49-49 McCain (was 47-52)

Rural: 53-44 Obama (was 48-50)

[Alcon]

Multnomah County, Ore.
New style:



Atlas style:

[Filuwaúrdjan]

What bist the key to the fust map [qm].

[Alcon]

What bist the key to the fust map [qm].

Same as here

[NOVA Green]

Hm, well, bump.

Overall, the model fared OK.  The biggest misses were pretty obvious weird results -- parts of Eastern Oregon, Hood River County.  Overall, it seemed to over-estimate Obama among downscale, culturally conservative oldschool Democrats and under-estimate him in the suburbs.  Then again, isn't that pretty much what happened everywhere?

(I knew Hood River looked ridiculous.  lol.)

Table


Map


-------------------------------------------------------------------------------------------------

So far, the only precinct results I have are preliminary ones for Polk County.  Since I'm bored, I'm posting them:

102 Ballston & Spring Valley: McCain +12 (was Bush +20) +8%
106 Perrydale & Buell:  McCain +12 (was Bush +23) +11%
110 Grand Ronde & Willamina: Obama +7 (was Bush +1) +8%
118 Rural Salem & Monmouth: McCain +8 (was Bush +15) +7%
122 Rural Independence: McCain +21 (was Bush +19) +2%
134 Salem City: Obama +1 (was Bush +10) +11%
144 Salem City: Obama +7 (was Bush +5) +12%
146 Salem City: Obama +5 (was Bush +7) +12%
152 Rural Dallas: McCain +17 (was Bush +25) +8%
154 Dallas City: McCain +7 (was Bush +17) +10%
162 Dallas City: McCain +5 (was Bush +19) +14%
166 Falls City & Surrounding: Obama +11 (was Kerry +8) +3%
178 Monmouth City: Obama +12 (was Kerry +2) +10%
182 Independence City: Obama +12 (was Bush +2) +14%

That's pretty boring from what I see, but I figure it might interest you as a local   Maybe you can gleam something out of Polk I couldn't.  (Other than the apparent lack of WOU electoral power.  Independence had a bigger swing than Monmouth?)

It is actually interesting...

West Salem is fairly conservative with a large number of retirees... Obama win significant and it helps explain Marion county to a small degree

Grande Ronde is partially an Indian reservation with a casino...

Dallas is the most conservative city in the county, decent former timber industry.

Fall City- outlaw country. Large number of counterculture and White rednecks and evangelicals.

Monmouth--- Western Oregon State University, as well as small HP commute population and some state workers in Salem

Indepdence-- decent Latino population

[Alcon]

If this isn't a stunning sign of the GOP's collapse in the Portland suburbs, I don't know what is:

Code:

          0352 352 BEAVERTON-CENTER

          United States President and Vice President
          Vote For  1
           Ralph Nader (PCE).  .  .  .  .  .  .  .  .      6    1.00
           Cynthia McKinney (PAC) .  .  .  .  .  .  .      2     .33
           John McCain (REP).  .  .  .  .  .  .  .  .    127   21.13
           Bob Barr (LIB).  .  .  .  .  .  .  .  .  .      8    1.33
           Chuck Baldwin (CON) .  .  .  .  .  .  .  .      2     .33
           Barack Obama (DEM)  .  .  .  .  .  .  .  .    450   74.88
           WRITE-IN.  .  .  .  .  .  .  .  .  .  .  .      6    1.00
              Over Votes .  .  .  .  .  .  .  .  .  .      0
             Under Votes .  .  .  .  .  .  .  .  .  .      3

Those sorts of numbers were replicated a number of places throughout the greater Beaverton area.  A few were even worse for McCain.

[NOVA Green]

If this isn't a stunning sign of the GOP's collapse in the Portland suburbs, I don't know what is:

Code:

          0352 352 BEAVERTON-CENTER

          United States President and Vice President
          Vote For  1
           Ralph Nader (PCE).  .  .  .  .  .  .  .  .      6    1.00
           Cynthia McKinney (PAC) .  .  .  .  .  .  .      2     .33
           John McCain (REP).  .  .  .  .  .  .  .  .    127   21.13
           Bob Barr (LIB).  .  .  .  .  .  .  .  .  .      8    1.33
           Chuck Baldwin (CON) .  .  .  .  .  .  .  .      2     .33
           Barack Obama (DEM)  .  .  .  .  .  .  .  .    450   74.88
           WRITE-IN.  .  .  .  .  .  .  .  .  .  .  .      6    1.00
              Over Votes .  .  .  .  .  .  .  .  .  .      0
             Under Votes .  .  .  .  .  .  .  .  .  .      3

Those sorts of numbers were replicated a number of places throughout the greater Beaverton area.  A few were even worse for McCain.

Holy Cow!

I wonder how much of this has to do with Latino support for Obama, since this area has a large immigrant population....

I have friends that live in central Beaverton, not too far from the Beaverton downtown mall. Heavily Latino, as well as a decent proportion of younger Whites and African-Americans. A Washington County map would be interesting if possible.

[NOVA Green]

Not sure if you have this link... looks like a pain to navigate since there are multiple maps of precincts for WashCo:

http://www.co.washington.or.us/deptmts/at/election/precinct/quadmaps.htm

[Alcon]

I'll need a GIS shapefile of Washington County.  I'll give them a try soon, but they may try to charge ridiculously.  It tends to be either free or that way.  I don't see the precinct file in their catalog.

Looking through Obama's performances: 80-16 Obama at Cedar Hils, 77-21 Obama at Sunset Cooridor (?!), 75-23 at Montclair, winning Cooper Mountain...

Basically it looks like Obama won every subdivision precinct in the county, and the only places McCain won were rural or, in a few rare cases, exurban (he barely won the McMansion sprawl part of Cooper Mountain.)  I'd really be interested in a map; it looks like the urban part of Washington County essentially Portlandified into solid red.  The only interruptions will probably be narrow McCain wins in Oregon City.