This page contains a single entry from the blog posted on November 7, 2008 8:24 AM. The previous post in this blog was Brown and Westlund were no Barack Obamas. The next post in this blog is Which 'dog will hunt?. Many more can be found on the main index page or by looking through the archives.

E-mail, Feeds, 'n' Stuff

Friday, November 7, 2008

If elections were horsehoes

What did I tell you in the wee small hours of Wednesday? Jeff Merkley won the U.S. Senate seat, and the suspense was because of the time it was taking to count votes in Multnomah County.

But what I wasn't able to predict with accuracy was the margin of victory. By the time they finished tallying up votes yesterday, Merkley had beaten incumbent Gordon Smith by 52,000 votes. On Wednesday afternoon, my projection was 15,000; and later on Wednesday, 35,000.

I did about the same on Measure 64, which I thought would fail by fewer than 10,000 votes. It wasn't as close as I thought, but it was close nonetheless -- fewer than 17,000 votes out of nearly 1.6 million.

This was my first attempt at amateur election-calling, and although I came up with a pretty good crude projection system -- looking at Multnomah County on the one hand, and the rest of the state on the other -- I see now that something a little more detailed is called for. What I should have done was go through the turnout and interim vote tallies on a county-by-county basis. And in these days of quick internet reporting of results and spreadsheet programs, that's actually doable from a blogger's armchair.

Next time there's a close statewide race, I hope to be able to say "I told you so" not only as to the winner but also as to the margin. For the wonkier out there, I may even post the spreadsheet as we go along with the count.

The internet age is an amazing time.

Comments (17)

I'm likely stuck with my Prophetic Poop Clouded-vision System. ... in case the internet breaks down.

I came to the same conclusion after KOIN reported a "startling" lead for Smith late Tuesday night. Even a quick look at the county-by-county totals would tell you that Merkley could make up the deficit with all the uncounted Multnomah votes. A more or less equal split in the 'burbs (Washington and Clackamas) plus a strong showing in Multnomah obliterates the rural vote. Sorry, Wheeler County.

It's mystifying why Gordo didn't campaign harder in Portland Metro. Dominating the Pendleton vote is not going to win a statewide race.

I was trying to track the Merkely-Smith race as a dabbling amateur, but found the SoS site to be quite unhelpful. By pretending that the votes counted at any one time was the same as ballots cast, it seemed very hard to discern the county-by-county progress at getting their ballots counted. Multnomah County distinguished the numbers, and it sounds like other counties did, too.

Are you thinking that you need to track progress on a county-by-county basis, or can we hope that Kate revamps the statewide website to give more accurate county-by-county numbers on the progress of the counting?

I was surprised Smith didn't run on his record. His TV spots were in poor taste. I thought Merkley was a weak candidate. I could have lived with Smith for another 6 years in the hopes a better replacement came along. Now we're probably stuck with Merkley for 24 years.

What record? What has Smith ever done to distinguish himself in the Senate?

Oh well, it's all over now and I suppose I should be nice. Next time a Republican runs for the Senate, he or she ought to accept the City Club debate invitation.

True, that may have been why he didn't run on his record. Too slim, but better than slime slinging. It seemed to work for Obama.

By pretending that the votes counted at any one time was the same as ballots cast, it seemed very hard to discern the county-by-county progress at getting their ballots counted.

There was another page besides the one you were looking it, and on that page the total turnout for each county was displayed as of late on Election Night. That page (pdf) is here:


Those numbers were set on Election Night and didn't change. From that and the returns themselves, you could see how far each county had gotten with their counting.

Ranking of US Senators

#100 Jeff Merkley

YEAH for Oregon

That's the way it works. Oregonians would rather have a junior Democrat than a senior Republican. Deal with it.

Did you get the feeling that Merkley just outworked Smith? Smith's campaign, and his presentation, just seemed really slack.

I didn't see Merkley outwork Smith at all, they were both slinging mud at the same pace. What I saw was an election year where pretty much all you had to do was put a "D" next to your name to win. Be honest, people are so mad at the "R's" right now they wanted them all tossed out on their collective rears.

Benschon.......No, Merkely definitely did not outwork Smith.

I got the feeling that folks in Multnomah County are so anti-republican, they would have voted for anyone other than Smith. The rest of the State was pretty split, so the folks in MC got to decide this one.
I believe that any democrat in Oregon could have beat Smith by 52,000 votes or even more.

This sentiment was true as well in other places too. If you don't believe me look how well an idiot like Al Franken did in Minnesota.

As someone observing your state from NC, I have to tell you we were really confused on election day. After we saw Smith ahead we could not understand why

1. The election was not called for him,

2. and then when we learned a lot of votes had to be counted, the question was why??

Of course, we are glad it turned out ok, but boy were we in the dark on this one.

In a previous comment I guessed that Minnesota recount would turn up vote tally-rigging against the rovian Coleman ... and so Al Franken wins.

So now this:

Most Minn. Senate 'Undervotes' Are From Obama Turf, By THE ASSOCIATED PRESS, November 8, 2008 .

ST. PAUL, Minn. (AP) -- An Associated Press analysis of votes in the tight, still-to-be decided race for a U.S. Senate seat in Minnesota shows that most ballots lacking a recorded choice in the election were cast in counties won by Democrat Barack Obama.

The finding could have implications for ... a recount separated by the thinnest of margins -- a couple of hundred votes, or about 0.01 percent.

About 25,000 ballots statewide carried votes for president but not for the Senate race. Although some voters might have intentionally bypassed the race, others might have mismarked their ballot, or optical scanning machines might have misread them.

A recount due to begin Nov. 19 will use manual inspection to detect such ballots.

Minnesota ballots are fed into optical scanners, which depend on voters filling in ovals to make their choice.

Recount teams will look for whether stray or light marks on ballots signaled a voter's preference.

The moral of the story is: papertrail ballots.

Also, there's that 'None of the Above' candidate(s).

Simple Sidney, Multnomah county is the largest county by population so the votes take longer to count. Also, the reason why the race was called for Merkley prior to the count is it is a prominent democrat stronghold. Very few republicans call Mult. co. home.

The reason Multnomah county took so long is that we have six machines for scanning ballots that can each scan 1,000 ballots per hour. So 360,000 ballots divided by 6 thousand per hour equals 60 hours to count all the ballots. Faster scanning machines would be great, but we've got what we've got.

Go by streetcar!

Clicky Web Analytics