Still Counting, Recounting and Runoff Compendium

by: Chris Bowers

Thu Nov 13, 2008 at 16:30

( - promoted by Chris Bowers)

With the Alaska Senate campaign turning heavily in favor of Democrat Mark Begich, in the extended entry I provide a run-down of the five closest campaigns that have still not been called, who is likely to win each campaign, and what it means for the overall balance of power. All of that, plus election forecasting notes can be found in the extended entry.
Chris Bowers :: Still Counting, Recounting and Runoff Compendium
Here is the best of what's left:

Election Run-offs

  • Georgia Senate: The run-off for Georgia Senate will take place on December 2nd. The only post-election poll on the campaign comes from Daily Kos / Research 2000, showing Republican Saxby Chambliss ahead 49%-46%. Turnout for the run-off will only be about 50%-60% November 4th levels. Also, it is unlikely that African-Americans, who were about two-thirds of Martin's voters on November 4th, will turnout at such high levels. So, the odds do not favor us on this one.

    However, Jim Martin's Act Blue pages are hopping, I am still running Personal Paid Media for this campaign:

    If you are running a Personal Paid Media campaign, or if you are simply interested in a form of search engine activism that is free, considering sending your ads to or embedding that website in a hyperlink whenever you write Saxby Chambliss. The site was created by a progressive activist who just noticed that Saxby Chambliss had left the URL wide open.

  • Louisiana 4th: Louisiana run-off elections will take pace on December 7th, making this the closing event of the 2008 campaign season. The district is R+7, but Democrat Paul Carmouche leads Republican Republican John Fleming by 10% according to an internal poll. Fleming did release a counter-internal poll showing him up 5%, but it has a hilariously small sample size of 140, and a margin of error over 8%. This campaign is somewhere between a toss-up and leaning Democratic.

Election Recounts

  • Minnesota Senate: Entering the recount, Republican incumbent Norm Coleman leads Al Franken by 206 votes. The vote will be certified on November 19th, and the recount will begin soon thereafter at 100+ sites across the state. Each campaign will have a lawyer at each site. After the recount, on December 16th, the canvassing board will begin to rule, one by one, on challenged ballots. Further, a fight over 461 disqualified absentee ballots in one county will go to court.

    So, this one is going to take a while. Franken will likely gain votes in the recount, but finding 207 will be difficult. The fact is that it is better to be ahead than behind in this situation. WineRev has a complete breakdown of the timeline and key players.

  • California 4th: Democrat Charlie Brown currently trails by 815 votes in this district. There are a few votes left to count, and a recount is also likely. However, the campaign is in a holding pattern until the first count is completed. This one looks unlikely.

  • Ohio 15th: Democrat Mary Jo Kilroy currently trails by 149 votes, with provisional ballots pending. There are thousands of provisional ballots, which always favor Democrats, and two years ago Kilroy gained more than 150 votes in that phase of the counting. This one still seems like a toss-up.

Balance of Power
There are a few other House seats that some consider undecided, but which I did not list here. While some are still waiting on AK-AL, I'm calling it for the Republican incumbent Don Young. Also, some are waiting on the LA-02 runoff with William Jefferson, but I have no doubt he will win. Finally, others are waiting on VA-05, but Tom Perriello will win there.

Overall, this puts the balance of partisan power in the U.S. House at D 257-175 R, with three campaigns outstanding. Either D 259-176 R or D 258-177 R appears to be the most likely final outcome, given that Republicans are favored in CA-04, Democrats are favored in LA-04, and OH-15 really is a toss-up.

In the Senate, I didn't list Alaska, but Mark Begich will win. This provides Democrats 56 seats plus Sanders and, as is appearing more likely all the time, plus Lieberman. Republicans right now only have 40 seats in the Senate, which is why I indicated last night that we are close to the point where Republicans don't matter anymore. Even though such a sentence smacks of hubris, I mostly stand by that assessment. It is accurate to say that if Franken and Martin both win, then Republicans won't really matter for pretty much all of 2009. So, we are close to the point where Republicans won't matter for a little while.

In the Presidential election, Obama currently leads the popular vote 52.69% to 46.01%. His margin has been steadily increasing over the last nine days, and will continue to do so until the final numbers are certified later this month. Missouri's eleven electoral votes are currently undecided, but Obama would need to make up 5,000 votes among provisional ballots. That seems unlikely, and the final electoral count will almost certainly be Obama 365-173 McCain.

That one electoral vote in Nebraska will really narrow down the final contestants for the Mac Book on Daily Kos's election contest. I won't be winning that, but I did win National Journal's contest for bloggers. That's something, I guess. Also, once final results are certified across all states, I will be interested to see how the various polling forecasting methodologies fared against one another. Specifically, I want to see how my method fared against fivethirtyeight and They both know more about polls and statistics than I do, but I wouldn't count my method out before the final numbers are in. I'm pretty sure there was virtually no difference in average error.

CQ Politics also has a good rundown of campaigns that are still undecided.

Tags: , , , , , , , , , (All Tags)
Print Friendly View Send As Email

Did I do it right? (0.00 / 0)
Testing my new tagline ...

Republicans can't fix our country; they're too busy saddlebacking.

Franken Will Win MN! (4.00 / 1)
Nate Silver on has a number of articles dissecting the electoral math and it all favors Franken who clearly is winning the recount battle.

Further, the undervotes will come largely from Hennepin County where Minneapolis-St. Paul is. That's Democratic country where Franken won by a wide margin.

An significant addition to the vote totals from Hennepin County will put Franken over the top.

Then we'll have the final vote certification, and after the Federal Court of Appeals rejects Coleman's inevitable court challenge that will be 58 Senators including Sanders, but excluding Lieber-thug.

It's likely at this point that Lieber-scum WILL stay in the caucus because of the total spinelessness of Democrats.

Then Lieberman will go on to be a TOTAL thorn in Obama's side for the next 4 years and NOBODY will dare do anything about it, because he's scream to the media that "Democrats are trying to silence my committee's investigations of Obama's unethical dealing in the [insert the scandal]."

He'll be MORE obnoxious than ever. Why wouldn't he? His ego  has to be sky high right now considering that he's totally gotten away with doing anything he wants. Democrats are all terrified of offending him and Republicans are kissing the ground he walks on. He's the media darling and everybody can kiss his ass.

I think it greatly amuses him that Obama is going to kiss and make up. Kissing up to Joe Lieberman in the expectation that "now he'll behave" is like expecting better behavior out Joseph Goebbels if you let him out of prison!

[ Parent ]
Wow... (4.00 / 1)
What a wildly inappropriate comment at the end...

Dude, we all hate Joementum... but seriously, how f-ed in the head are you to compare a US senator to a Nazi.... much less a JEWISH US Senator to a Nazi.    

There were 50 better analogies to make and you picked the one that made you look like a huge asshole.  

Well done.    

[ Parent ]
Fun and yes it worked. (0.00 / 0)
Although, and Im not sure it's best, though it is most effective moving forward. It doesnt just change your sig for new posts, it alters all your old posts too.

So If you go look at your post from June when you where telling (Obamanauts to relax or Hillbots to be kinder)or something, you'll find you were already exposing Saxby Chambliss.


The government has a defect: it's potentially democratic. Corporations have no defect: they're pure tyrannies. -Chomsky

[ Parent ]
Yay! (0.00 / 0)
I'm happy to retroactively diss Saxby Chambliss ...

Republicans can't fix our country; they're too busy saddlebacking.

[ Parent ]
Re. MN Recount (0.00 / 0)
How many votes did Gregoire trail by in the 2004 Washington Governor's race before she pulled ahead in the recount? Who led prior to the 2000 senate recount?

OK, here it is (0.00 / 0)
Rossi led Gregoire by 261 votes after the initial count. Cantwell led by over 2,200 votes in 2000.

[ Parent ]
Cantwell 2000 (0.00 / 0)
There wasn't a recount in 2000.  Slade Gorton, the incumbent, was ahead in the counted votes in the days after the election. But vote-counting was extremely slow (as often in Washington) and Cantwell came into the lead at the very end, 10-14 days after the election, when the last Whatcom county votes were added to the totals.  

[ Parent ]
There was a recount (0.00 / 0)
according to Wikipedia. Cantwell increased her lead by a couple hundred points.

[ Parent ]
I did some calculations... (4.00 / 1)
In 2000, Cantwell gained 276 votes between the initial and final results, out of approx.* 2396645 votes. In 2004, Gregoire gained 390 votes out of approx. 2746593. Franken is currently down by 206 out of approx. 2422924** (with, I believe, the county with Minneapolis in it still yet to finalize their results?). If he gains votes at the same rates as Cantwell and Gregoire, he is likely to gain between 279 and 344 votes. (Caveat: Washington is not Minnesota. Their voting machines are likely different, their voting and counting procedures are likely different, their laws are likely different, and so on. So they're not directly comparable, but it's an interesting excercise.)

* I say approximately because I didn't make sure to use either the initial or final vote totals -- I'm not sure which is statistically correct, and I used whichever I found first. The difference is likely to be miniscule.

** In each case, I added together the total votes of Democratic and Republican candidates, because I didn't find a sum of vote totals anywhere on Wikipedia. Minnesota had a much stronger third party challenger than either of the races in Washington, so Franken's estimated gains should probably be proportionally even higher.

[ Parent ]
Wouldn't 3rd party make it lower? (0.00 / 0)
[blockquote]Minnesota had a much stronger third party challenger than either of the races in Washington, so Franken's estimated gains should probably be proportionally even higher.[/blockquote]

All other things being equal, shouldn't the strong third party make Franken's estimated gains LOWER?  Perhaps you're right that it would be proportionally higher, but lower in terms of absolute votes.  when you need to make up 206 votes though, the absolute votes are what really matters.

Let's say there are 10,000 (rather random number) "new votes" to be found in the recount.  In a 2 way race each person should be expected to get around 5,000.  In Minnesota this year we should expect Franken and Coleman to get around 4,200 votes while Barkley gets around 1,500.  Thus Franken will need a higher % advantage in the recount than the folks in Washington needed.

[ Parent ]
On the other hand, (0.00 / 0)
the more total votes, the more "missing" votes you should expect to find. (Rate of missing votes) * (D+R vote) isn't significantly different from (Rate of missing votes) * (D+R+I vote) when the Independent vote is like 2%, but when it's 16% it's a sizeable bump. But the point you make is also a good one. So maybe the method I used of only counting the D+R vote is actually the best approximation.

[ Parent ]
It's interesting to look at polling for states (4.00 / 1)
As the numbers have come in, most states have been in the margin of error, while some states weren't.

Example - Nevada - supposedly a swing state, right? - went BY 12 to Obama.

That wasn't the polling average there, but instead was a big outperformance by Obama.

To the extent there was a miss by polling in the states, on average, the blue states tended to overperform towards Obama, while the red states tended to overperform for McCain.

It's also been interesting seeing the margins for blue states, continually creep up, in this last week.  Virginia ended up +6, Ohio ended up +4, Colorado +9.  Some nice margins to build on, in the supposed "swing" states, going forward. noticed this too. (0.00 / 0)
Sam Wang (Princeton Academic specializing in statistics and accomplished amateur poll watcher -- it looks like he may have predicted things better than Nate Silver this time around) has an interesting post with some insightful comments about this effect here:

[ Parent ]
Re: GA-SEN (4.00 / 2)
Obama's still got this huge email list and generally this  entire monster turnout operation from about two weeks ago still basically mostly functioning and potentially able to be fired up again, right? Is it planned for any of the Obama campaign infrastructure to be used to encourage GOTY in the GA runoff?

Dairy on DailyKos (4.00 / 2)
said that all of Obama's Georgia staff have been lent to Martin (can't find the link right now, but it sounded promising).

I haven't seen a list-wide email, but maybe he sent something targeted to Georgia? Any Georgia readers got anything?

[ Parent ]
I volunteered in the Obama campaign (4.00 / 2)
here in Macon for the last 2 months of the campaign.  The paid staff left lask week to resume their lives.  At the election celebration on the night of the election, there was an African American who was taking names of attendees who would work for Martin.  I also volunteered with the Martin campaign at its website.  I contributed via Act Blue.

No one has YET contacted me about working.  I know, because I phone banked and physically canvassed, that Martin will win only if a substantial portion of the African American vote turns out (Obama carried my county with 60% of the vote and a 77% turnout of registered African American voters).  Perhaps the campaign is focused in the Atlanta area, but if I were Martin, I'd focus on the Fulton/Dekalb portions of Atlanta, Clarke (Athens), Muscogee (Columbus), Bibb (Macon), Chatham (Savannah), Doughtery (Albany), Richmond (Augusta) and a few others.  Actually, Martin's vote difference with Chambliss was less than that of Obama versus McCain, although it was clearly Obama's strength with AA"s that got Martin as close as he was on Nov. 4.

 Saxby has already brought out the big guns, and we have seen nothing but TV ads from Martin, who has to be tired.  He has been in the primary in July, then a runoff for that in August, then in the general in November and a runoff for that one.  He was and is a virtual unknown in the state and actually out there with the media.  Even on Olberman and Maddow, the talk is about Chambliss.  The name of Martin sometimes doesn't even get mentioned.  Not good for the home team.  We need a real attention getter.  As good a guy as Martin is, as the AJC says, charisma is not his strong suit.  Martin is from the Atlanta area.

[ Parent ]
Has Jim Marshall come out strongly in favor of Martin? (4.00 / 1)
Would be nice to see the GA Blue Dogs pushing the Dem Senator over a real piece of work like Chambliss.  Thought you might know, since you're from Macon...

[ Parent ]
Dailykos thinks that there's still a good chance in CA-04 (0.00 / 0)
Supposedly, about 36k ballots left, most from the pro-Brown Broward county.

Saxby Chambliss  

Broward? (0.00 / 0)
There is no such county in CA.  Perhaps Placer or Nevada?

[ Parent ]

Open Left Campaigns



Advanced Search

Powered by: SoapBlox