More still on the DeFazio/Smith poll

Charlie Burr

Here’s my nomination for most bizarre analysis of the 2008 U.S. Senate race to date:

DeFazio's populism has carried him to 11 terms in Southwestern Oregon, but he's never demonstrated he could carry the Democratic-leaning Portland metro area.

That nugget of wisdom comes from Diane Dietz in today's Register-Guard. If Portland is only "Democratic-leaning", as Dietz writes, it begs the question of what a Democratic stronghold looks like.

As a reality check, it has been nearly a quarter century since a Republican Presidential candidate carried Portland. While Republicans have made gains in statewide registration, Portland Democratic performance continues to trend decidedly blue. In fact, last year’s gubernatorial race saw Portland-based attorney Ron Saxton lose Multnomah County by a nearly three-to-one margin, despite his candidacy being largely premised on a strategy of cutting Republican losses there. It's hardly an exaggeration to say that Dennis Kucinich will be elected President of the United States before DeFazio loses Portland to Gordon Smith.

To reinforce the Portland-as-swing-area statement, Dietz quotes political analyst Jim Moore. From the piece:

Moore cites DeFazio's loss to Democrat Ron Wyden in the senatorial primary in 1995. "He really had a hard time generating buzz outside of Eugene," Moore said. "He just got creamed."

This analysis is flawed in several ways. First, performance in a Democratic primary election 11 years ago is hardly a useful measure for a general election contest in 2008. Second, to write that DeFazio got “creamed” is revisionist history: DeFazio earned 47 percent of the vote, hardly the blow-out Dietz reports. Third, this analysis is clearly contradicted by the poll buried in the story. There is no way DeFazio is leading Smith statewide without enjoying a lead in Portland today, even before declaring his candidacy.

More from the Dietz piece:

Further, Moore said, "People, by and large, tend to trust Gordon Smith or they say `Isn't it good that we have one D and one R in this day and age when partisanship is a problem? And they (Wyden and Smith) work so well together.'

That's remarkable analysis for a story that reports Smith trailing more than a year before the election. Here's how I read a poll in which an incumbent is losing to a challenger yet to announce his candidacy: voters are VERY open to change and there are many ways and candidates who could potentially defeat Smith.

Peter DeFazio would make an extremely strong candidate. But this poll, mentioned only once in the eighth paragraph, arguably tells us even more about Smith's vulnerability than DeFazio's strength. Both Dietz and Moore reach exactly the wrong conclusions.

  • verasoie (unverified)

    I have a hard time believing that DeFazio couldn't raise a ton of cash from the grassroots here in Oregon. The energy behind his run would be amazing.

    You wonder if the DSCC money comes with some strings attached, though they don't know who they're dealing with if they think they can own DeFazio-- he would vote his conscience every time and not bow to them if (when) elected.

    Go DeFaz!

  • (Show?)

    Barack Obama has also "never demonstrated he could carry" New Hampshire either. What a stupid bit of analysis, Diane.

    Yeah, um, DeFazio represents Oregon's FOURTH district - which doesn't include Portland. It's not like he's been trying, year after year, to win votes in Portland and goshdarnit just can't seem to pull it off.


  • Peter Bray (unverified)

    Well, Bradbury captured 58% of the vote in Multnomah County against Smith (and other right-leaning candidates) in 2002. Smith of course went on to win.

    Bradbury doubtlessly has better name recognition than DeFazio, so while Smith would doubtlessly lose Portland, it wouldn't necessarily be by a long shot.

    Can DeFazio beat Bradburys 58%? That's the real question. I would guess: no.

  • Charlie Burr (unverified)

    A few more points on the piece:

    Deitz fails to mention the actual numbers -- 42 for DeFazio, 38 for Smith (I excluded them only b/c we've covered the poll extensively on previous posts).

    Dietz omits the context of the Democratic primary referenced -- that downstate Congressman DeFazio was facing off against Ron Wyden, at that time a beloved Congressman from Portland.

    It also appears unlikely that Deitz attempted to contact pollster Lisa Grove for a response. While I understand that this is a DSCC-sponsored poll, Grove is among the best in the country and has an extensive record of Oregon wins backing up her market-based research.

  • pedro (unverified)

    what idiotic analysis.

    oh, and, uh... what does this have to do with obama and new hampshire???

  • (Show?)

    In the RG piece, Diane argued that Peter hasn't demonstrated success winning votes in Portland - even though he's never been in a general election that included Portland.

    That's exactly like claiming that Obama can't win in NH, because he's never won votes there before. Of course he can win NH.

    And Peter can win in Portland. Duh.

  • YoungOregonVoter (unverified)

    Really flawed analysis.

    The basic assumption is that the Portland Metro Area is the whole state and that on a electoral map, the Portland Metro Area has and will always be 100% blue.

    Facts Clackamas County - 361,000 Multnomah - 692,825 Washington - 489,785 Total of 3 - 1,543,610 Oregon Total Pop.- 3,631,440 Big 3s % of pop. - 42.5%


    Knowing that out of all the voters in Clackamas, Multnomah, and Washington there is healthy percentage (around 20% to 33% in each county) of Republican voters, then this analysis becomes more flawed.

    On the rural side (outside of Clackamas, Multnomah and Washington), there are blue counties (Lane County) and in the red counties such as Umatilla there is still a healthy percentage of Democrat voters.

    It is not as simple as saying that whoever carries the Portland Metro will win the 2008 election. There are many complexities at all levels with all voters.

  • YoungOregonVoter (unverified)

    My last post was also working from the assumption that if a voter is Democrat then they will vote Democrat. Likewise, if someone is registered Republican, then they will vote Republican.

    This is an overly simplistic, unsubstantiated assumption.

    Trying to control for independent voters and those registered Democrats and Republicans whose vote is based entirely on the candidate and his/her character or positions regardless of party affiliation is extremely difficult if not impossible knowing the independent streak of American voters.

    I apologize for rambling too long.

  • (Show?)
    Bradbury doubtlessly has better name recognition than DeFazio...

    not necessarily. Bradbury is only Secretary of State, and i wouldn't be surprised if half the voters in this state have no idea who he is. DeFazio may not be known to all voters, but you add "US Representative" or "Congressman" before his name, and people immediately "understand" that he is, of course, highly qualified. it's not like we're dealing with members of Harvard's School of Government here; we're dealing with voters who spend about 6 nanoseconds a day thinking about politics. that's our big problem: they pay so little attention, they have no idea what suckers Smith plays them for -- and what a great Congressman Peter DeFazio is.

    in the end, no matter who runs, it's going to take a tremendous grassroots effort to beat Smith. he'll have enough money to win a conventional, media-based election; we have to expand the field of conflict greatly.

  • (Show?)

    Portland is not a swing area. That doesn't mean that a strong metro field and GOTV effort isn't needed -- or that the rest of the state should be ignored. Of course it shouldn't. But what Dietz writes -- that there's any real question about DeFazio "carrying" Portland against Smith -- is contradicted by years of election results.

    I don't think anyone underestimates Smith, btw. He's got formidable political skills, including the ability to not lose by the same large margins in Multnomah county that most Republicans do. That's how he's done well in the past, but I think 2008 will be different.

    To Peter's question, I do believe DeFazio 08 would outperform Bradbury 02, or for that matter so would a theoretical Bradbury 08 candidacy. In 2002, Bradbury had little resources to respond to Smith's attacks. More importantly, 2008 will be a Presidential year with a strong field effort for the inevitable Dem Prez nominee. It's also worth noting not just how scarce Bradbury's resources were back then, but also how different the political climate was -- especially on the war.

  • (Show?)

    I happened to be in Eugene and read this story. What a badly thought through or out-and-out slanted piece of analysis. If Steve Novick runs, I am supporting him, but of course if DeFazio is the nominee I will work my ass off for him. And I'm confident that Portland will NOT be a problem in that race.

    Jim Moore is a partisan Republican pollster, n'est-ce pas? Why did the Register-Guard leave that nugget out?

  • Anon (unverified)

    Bear, I don't disagree with your larger point about the R-G here, but I think you're thinking of Republican pollster Bob Moore, not Pacific University professor Jim Moore quoted extensively in the article.

  • (Show?)

    Oh, I'm very sorry. Good catch. I was thinking of Bob Moore of Moore Information.

  • (Show?)

    If DeFazio were our nominee, he could focus heavily in swing counties and more rural ones within his own OR-04 district, that will cut into Smith's "safe" areas. As this post indicates, DeFazio wouldn't have to fight hard to blow Smith out in Portland Metro. DeFazio can focus on Clackamas and other swing and slight-R leaning areas, where his populist record and approach can do serious damage to Smith's fake moderation appeal. Smith is going to have to fight DeFazio in such areas,m which means Smith is not focusing on Portland metro/suburbs which is where he has had to cut into suburban votes to win.

    A candidate that can at least make a respectable showing (not win mind you) in places like non-Eugene OR-04 and in Clackamas means that Smith would have to be focusing on his own back-yard instead of playing offense in Portland suburbs.

  • (Show?)
    Posted by: Peter Bray | Apr 3, 2007 10:17:22 PM

    You are smoking something for medical purposes there? (wry grin)

    DeFazio would do far better than Bradbury (who I like, but was/is a very lackluster campigner). Portland metro wants Smith GONE. DeFazio would be doing respectable in places where Bradbury couldn't [and force] Smith to [defend traditional R] areas instead of trying to mine Portland suburban votes.

    [Editor: comments lightly edited for clarity]

  • (Show?)

    Whoa, what an incredibly poor article. And how can a politics professor be so clueless? Moore's analysis is shockingly flawed. Dunno what's going on out there at Pacific, but it's too bad Russ Dondero retired...

  • Trish Binder (unverified)

    I tried to post earlier but it doesn't seem to have gone through.

    DeFazio has the advantage of being strong in rural areas, unlike many Democrats. His office also has the best constituency support of any politician I know of in Oregon. He won't have to rely on urban support alone.

  • (Show?)

    Charlie: Portland is not a swing area. That doesn't mean that a strong metro field and GOTV effort isn't needed -- or that the rest of the state should be ignored. Of course it shouldn't. But what Dietz writes -- that there's any real question about DeFazio "carrying" Portland against Smith -- is contradicted by years of election results.

    Didn't Smith win the PDX tri-county region against Bradbury in 2002? Also worth considering is that Smith has strong favorables among Democrats.

    I'm not saying that you're wrong, Charlie, only that there's a reasonable case to be made for some of the concerns about how well Rep DeFazio will do in the Portland Metro area.

    Peter Bray: Can DeFazio beat Bradburys 58%? That's the real question. I would guess: no.

    Perhaps, Peter, but let's not forget that Multnomah does not comprise the entire Portland metro area. I suspect that DeFazio will do better in Washington County and Clackamas than did BB.

  • (Show?)
    Didn't Smith win the PDX tri-county region against Bradbury in 2002? Also worth considering is that Smith has strong favorables among Democrats.

    No, Smith did solidly but didn't win outright there, but that was 2002, not 2008. He had some favorables among Dems in 2002 (by bamboozling and pretending to be moderate on things like hate crimes law), but HAD is the operative word, not has. Smith is vulnerable. He will have next to no support in Portland metro this time, and should DeFazio be the Dem candidate, he can do the exact opposite of what garnered Smith's win in 02, by DeFazio doing solidly in rural areas of OR-04 and make Smith have to fight to hold ground there (and thus not be engaged in trying to re-secure his lost creds in the Portland burbs (Washington county specifically).

  • (Show?)


    According to the official results from the SOS, Smith won Clackamas and Washington Counties by a combined 60,000+ votes and lost Multnomah by 41,000 votes.

    My reference to Smith's favorables among Democrats were based on current polling that was posted here. Smith has a 62% approval rating among D's, according to SurveyUSA:

    I take all of that with a grain of salt, because I do believe that Rep. DeFazio will beat Smith if he decides to run.

  • (Show?)
    Posted by: Sal Peralta | Apr 4, 2007 3:57:20 PM

    Thanks for the links, I stand corrected about Smith's performance in Washington county in 2002. Somehow I was mixing it up in my head with the results in WA county in Gov. race last year.

    Survey USA's polling is dubioous at best (they had close to the worst record in last years cycle). Other polling has Smith's fav/unfav in the mid 50 favs vs almost 40 unfav

  • (Show?)
    Posted by: Sal Peralta | Apr 4, 2007 3:57:20 PM

    FYI, Survey USA has more recent approval polling on Smith and it more damning against Smith than the one you cited:

    GOP 56/36
    Dem 46/45

    With an overall of


    That said, I still take Survey USA with some granules of sodium chloride. (grin)

  • Charlie Burr (unverified)

    Sal, you're point's well taken about the underfunded Bradbury campaign's vote totals in Washington and Clackamas counties. But the Grove poll -- more reliable and accurate than Survey USA -- pretty much puts to bed any serious question about DeFazio's ability to carry the Portland area.

    Dietz underestimates Democratic performance. For example, Dietz called the Kulongoski/Saxton race last year a "squeaker" just days before the election (Register-Guard, 11/1/07). Again, that wasn't in January when many considered the race wide open, it was during a time when every opinion poll and every political pundit in the state had basically called it for the governor. Compare Dietz's "analysis" with the actual vote totals days later.

    More importantly, the AP version of this story omitted the flawed Dietz "analysis" and included the one candidate who's expressed genuine, unabashed enthusiasm for running, Steve Novick. Both DeFazio and Blumenaur would make great candidates, but there are many ways -- and strong potential candidates -- to defeat Senator Smith. I don't underestimate Gordon's political skills, but believe with the right unorthodox campaign anything can happen. The political climate's ripe for an upset. I assume a close race no matter who runs, so am looking at intensity and who has the potential to catch fire. That's not a list limited to our talented Congressional delegation.

connect with blueoregon