My Election Calculator

Forget Blanchard's election shaman.  He has no more than rolled bones and chicken entrails to guide him.  I have science. 

A couple weeks ago I composed a blog post making assumptions about the Kristi Noem vs. Stephanie Herseth Sandlin (SHS) race based on voter registration numbers, assumptions about turnout, and assumptions about the likely allocation of votes between the two candidates by Republicans, Democrats, and Independents.  The KELO poll allows us to revisit those assumptions with a bit more rigor. 

According to the Secretary of State site, as of October 1, 2010, voter registration in the state was this: 235,906 Republicans, 193,304 Democrats, and 83,799 Independents (I will leave out the minor parties as they aren’t significant numbers).  Here is my big assumption: Looking at the turnout data for a number of years, it looks like a typical mid-term election in South Dakota yields a turnout of about 70% of registered voters.  Judging by trends I see, I am assuming that Republican turnout will be a little higher; I am going with 75%.  Democratic enthusiasm seems to be lower; I’ll go with 65% turnout.  I’ll put Independents in the middle with 70%. 

Looking at the KELO poll we see that SHS wins the Democrat vote 81% to 11%, with Independent candidate B.T. Marking getting 2%.  Allocating the undecided voters proportionally to those who have decided, we can guess that SHS wins Democrats 87% to 11% to 2%.  Noem is winning the Republican vote 70% to 17% with Marking getting 3%. Again, allocating the undecided vote proportionally, I estimate that Noem will win 77% of Republicans, with SHS getting 20% and Marking 3%.  Independents end up this way after allocating undecided voters: Noem with 49%, SHS with 39% and Marking with 12%. 

 When one does the math using my turnout assumptions, one gets this result: Noem 178,799 votes, SHS 167,575 votes, and Marking 14,859 votes.  That’s an 11,000 vote win by Noem.  Assume I underestimated the Democrat turnout.  Say it is 70%.  That adds another 10,000 Democrat voters.  This is not enough to overcome the gap between Noem and SHS even if all those voters go to Herseth Sandlin.  Of course I could be overestimating Independent and Republican turnout.  Well, you play with those numbers.  

What should really frighten Herseth Sandlin (again, if the poll is accurate) is that she is getting soundly beaten among Independents.  This is where she has long had a big advantage.  Without going through the math, I figure that in 2006 and 2008, SHS won 50% of the combined Independent and Republican vote (it is impossible to disaggregate the two given the Secretary of State’s data).  That tells me the SHS has been winning a super-majority of Independents.  She is unlikely this year to even win a majority, according to the KELO poll.  I am also assuming BT Marking gets 12% of the Independent vote.  I actually suspect that won’t occur.  I think he will shed some of his support over the last week and those voters are more likely to go with Noem over the incumbent. 

So here is my guess.  Noem  gets 51%, Herseth Sandlin 47%, and Marking 2%.  Kristi Noem wins the race.  If I am wrong and Herseth Sandlin wins, I predict that it is by fewer than 2,000 votes, which is less than 1%. 

For what it's worth, I think Republicans will win just north of 50 House seats and 7-8 Senate seats.  I also predict that John Thune wins in a nail biter. 

Comments

[Bill Fleming]

This poll from the Hill has it thus:
SHS 45%, Noem 42%, 10% Undecided SHS winning Indeps by 12 points.
http://thehill.com/house-polls/thehill-poll-week-4/125979-district-by-district-south-dakota

A reliable source tells me that both campaign's internal polls agree
and contain results more closely resembling the Hill poll than the Argus Poll.

As usual, the reservation vote hasn't been sampled in either poll,
nor, I'm told, have people who use cell phones exclusively.

The kicker is that women and the older demo (55+) prefer SHS.

I agree, it will be a close finish. But I disagree on the Marking number.

I think he might pull 5%. Just a hunch. (I am my own shaman.)

Here's my call: SHS 48%, Noem, 47%, Marking 5%

[Jon S.]

Bill,

I think if Herseth does win the numbers will look a lot like what you are saying.

[William]

My prediction (worth what you pay for it)

Noem = 52%
Marking = 3%
Herseth Sandlin = 45%

[KB]

My Shaman is sowing a doll that looks remarkably like Professor Schaff, only it's not quite as ugly. Maybe the magic won't work.

[KB]

My Shaman is sowing a doll that looks remarkably like Professor Schaff, only it's not quite as ugly. Maybe the magic won't work.

[George Mason]

Professor Schaff; You are really out on a limb with your Thune victory prediction. I guess we'll know when the final results are in.

[Dammika]

Is this even relevant? It's not like annoye stood to gain from this article, because it was poorly written and lacks any other back round information about these two senators and the reasons on why they left.If Ellefson missed a meeting for his ailing grand mother than I'd expect the SGA president to take him for his word, not assume foul play and deny the proxy until there was factual evidence that proves the drive to Pipestone did not happen. Trips don't always run late, people may arrive back sooner than expected, but even in he got back early enough, why stop him from attending a political organized event? Did Hughes even ask Ellefson why he bolted from York? That would probably have benefitted the reader more than anything, and proves that after writing a biased and horrible article that this writer needs to go back to Intro to Journalism.(I am definitely not journalism major, let alone English)This article has a place and time, solely when it's a complete article which this is not!There are a lot of holes in this article and i'm sure A LOT of the back round information left out. Clean up your work Collegians!

[Culii]

Where on Obama s promised line- by-line nonaital budget review do we insert this illusion of refugee care? Next to the social services costs for the15 million Americans that cannot secure a job to feed their own families as well as millions suffering foreclosures on their homes? Note: U.S. taxpayers stand $13 trillion in debt today and counting! Meanwhile, let me repeat: third world countries add, births over deaths, a mind-numbing 80 million fresh and precious, yet starving, poor and uneducated babies to their countries every year. Result: no end of the line for refugees that race into the United States for a better life. Within 60 years, less than the lifetime of a child born today, this immigrant overload will grow the US population from 310 million to over 600 million. By that time, our problems as to water, energy, environment, pollution, species extinction, acidified oceans, collapsing fisheries, climate destabilization, dwindling resources and lowered standard of living, and quality of life will pretty much resemble present day China, India and Mexico.

[gera]

Can you guess which organization is gultiy of the following??? 36 have been accused of spousal abuse 7 have been arrested for fraud 19 have been accused of writing bad checks 117 have directly or indirectly bankrupted at least 2 businesses 3 have done time for assault 71 repeat 71 cannot get a credit card due to bad credit 14 have been arrested on drug-related charges 8 have been arrested for shoplifting 21 currently are defendants in lawsuits,and 84 have been arrested for drunk drivingin the last year It's the 535 members of theUnited States Congress! The same group of Idiots that crank out hundreds of new laws each year designed to keep the rest of us in line. Remember, Election Day is coming!!