Senate '08: What to watch for in early returns

Kari Chisholm FacebookTwitterWebsite

Watching the returns each election, there's always that weird moment where you ask yourself: "OK, we're doing well in that county - but how well are we supposed to do there?"

Or, put another way, if a Democrat gets 60% in Multnomah County - is that good news or bad news? (Hint: it's terrible news.)

So, I thought I'd look back at some historical elections to see if I could identify a handful of counties that could give us an early read on how Jeff Merkley and Gordon Smith are faring.

I looked back at US Senate, Governor, and Presidential races in Oregon - going back to 1996. (Over that period, Democrats won every one of these top-ticket Oregon contests except the two campaigns against Gordon Smith, by Bill Bradbury in 2002 and Tom Bruggere in 2006.)

Here's what I found:

Multnomah County: When Democrats win, they win at least 69% of the vote in Multnomah County. Kulongoski '06 and Kerry '04 each picked up 73% - while Bruggere '96 and Bradbury '02 got 64% and 59% respectively. If the early returns from Multnomah County have Smith at 30% or more, we're in trouble. (A point made by Jeff Mapes in his analysis.)

Washington County: Another strong correlation here. Bruggere got 47%. Bradbury got 39%. But every winning Democrat has won more votes in Washington County than the Republican. (Well, except Kulongoski '02, which was basically a draw - 49.7% to 50.3% for Mannix.) As David Sirota has noted, if Jeff Merkley performs well, say 50% or so, in the pro-free-trade and high-tech suburbs, we're in great shape. If he does substantially better than that, say 53 or 54%, he's in a commanding position.

Marion County: This mid-valley county has long been a swing region in our state. Bradbury was held to 35% in 2002, while Bruggere only won 42%. Every winning Democrat has won at least 45% of the vote here.

Lane County: While centered on liberal Eugene, Lane is actually a very big and rural county. Nonetheless, it's critical that Jeff Merkley piles up some big numbers here to win. If he's at 59 or 60% (like Kerry '04 or Kulongoski '02), he's in great shape. 55-56% is fair - that's what Wyden '96 and Gore '00 did, as well as Bruggere '96. Under 50% like Bradbury '02? Merkley's in trouble. (Remember that Lane County was where Merkley won the primary with a huge 11-point win.)

Jackson County: Down south, Democrats have been doing an amazing job getting organized. And it's definitely a key swing region that includes cultural opposites Ashland and Medford. If Merkley picks up 40-42%, the race could go either way (both Bradbury '02 and Wyden '96 had only 40%). But if Merkley climbs up to 44, 45, 46, 47 (as Kulongoski did in 2002), then he's looking good.

Obviously, there's a lot more number-crunching and analysis that could be done with this data. And certainly, all the best-case scenarios above could happen and Jeff could still lose - if there was a complete meltdown in some other counties. But I'll be personally watching these five counties to see if it's time to put the champagne on ice or start breaking out the tequila.

A caveat: All the numbers above are calculated as a two-candidate percentages; for consistency across elections, I ignored the minor party candidates.

  • (Show?)

    [Full disclosure: My firm built Jeff Merkley's website, but I speak only for myself.]

  • (Show?)

    One thing to consider that should skew the numbers is that in the recent past elections you've looked at, the voter registration differential between the two partie was much smaller than it is today, so the margins in traditional Democratic strongholds need not be as strong as in year's past because the margins in counties where Democrats typically do not perform well has shrunk significantly heading into this election cycle.

    For example, in Yamhill County during 2004 there was a +4,600 voter registration edge for Republicans county-wide out of 47,00 voters. In 2008, that number is +1,700 for Republicans out of 50,000 registered voters.

    In Marion County, we've seen a +8,000 registration edge for Republicans in 2004 turn into a +2,000 registration edge for Democrats in 2008.

    It's a pattern that we've seen all across the state, and I wonder what the longer-term implications are.

    YEAR    DEM     REP 
    2008    936,735     697,309
    2006    763,960     700,950
    2004    820,602     757,844
    2002    726,187     680,650
  • (Show?)

    Jeff could dip below 70 percent and still pull off the race, but it'd certainly make for a long night (and next day). In 2004, we won the medical malpractice initiative with about 63 percent of the Multnomah County vote, with the deficit made up in Washington County plus other traditional left-leaning pockets.

  • (Show?)

    Sal -- You're absolutely right. My entire analysis was based on the percentages within counties - while the relative size of the various counties has shifted, both in total and in partisan makeup. And since Democrats are turning out in larger numbers, we can presume that Democratic-leaning counties will have larger turnout.

    And Charlie, you're right too. I left ballot measures out of my analysis because the breakdown doesn't always break along the same partisan lines.

    There are a zillion scenarios under which my best case scenario for those five counties could come true and Merkley could still lose -- or under which my worst case scenario happens and he could still win.

    As they say in football, that's why you play the game.

  • Scott J (unverified)

    How many new Dem voter registrations this year was due to Republicans changing registration to vote for Hil Clinton as part of Rush's "Operation Chaos"?

    Is there any estimate of this?

  • (Show?)
    <h2>Thanks for this, Kari...this is why you get the big BlueOregon bucks....oh, wait. Nevermind.</h2>

connect with blueoregon